Progress in Oceanography 131 (2015) 8299

Contents lists available at ScienceDirect

Progress in Oceanography
journal homepage: www.elsevier.com/locate/pocean

A real-time PCR assay to estimate invertebrate and sh predation

on anchovy eggs in the Bay of Biscay
Aitor Albaina a,, Xabier Irigoien b, Unai Aldalur a, Unai Cotano c, Mara Santos c, Guillermo Boyra c,
Andone Estonba a
a
b
c

Laboratory of Genetics, Dpt. Genetics, Physical Anthropology & Animal Physiology, University of the Basque Country UPV/EHU, Leioa 48940, Spain
King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Saudi Arabia
AZTI Tecnalia, Marine Research Division, Herrera Kaia Portualdea z/g, P.O. Box 20110, Pasaia, Gipuzkoa, Spain

a r t i c l e

i n f o

Article history:
Received 14 July 2014
Received in revised form 1 December 2014
Accepted 1 December 2014
Available online 12 December 2014

a b s t r a c t
In order to investigate the role of predation on eggs and larvae in the recruitment of anchovy (Engraulis
encrasicolus), sardine (Sardina pilchardus), sprat (Sprattus sprattus) and 52 macrozooplankton taxa were
assayed for anchovy remains in the gut during the 2010 spawning season using a molecular method. This
real-time PCR based assay was capable of detecting 0.005 ng of anchovy DNA (roughly 1/100 of a single
egg assay) in a reliable way and allowed detecting predation events up to 6 h after ingestion by small zooplankton taxa. A total of 1069 macrozooplankton individuals, 237 sardines and 213 sprats were tested.
Both sh species and 32 macrozooplankton taxa showed remains of anchovy DNA within their stomach
contents. The two main ndings are (1) that the previously neglected macrozooplankton impact in
anchovy eggs/larvae mortality is in the same order of magnitude of that due to planktivorous shes
and that, (2) the predation pressure was notably different in the two main spawning centers of Bay of
Biscay anchovy. While relatively low mortality rates were recorded at the shelf-break spawning center,
a higher predation pressure from both sh and macrozooplankton was exerted at the shelf one.
2014 Elsevier Ltd. All rights reserved.

Introduction
It is generally accepted that much of the marine sh year-class
strength variation is determined by early life stages (ELS) mortality
(e.g. Houde, 1987; Bailey and Houde, 1989; Bailey, 1994; Bax,
1998; Bunn et al., 2000; North et al., 2009). Between others (starvation, transport away from nursery grounds, disease, parasitism
and pollutants) predation is considered the main cause of natural
mortality of marine sh ELS. Mortality rates for temperate species
ELS are between 5% and 20% per day (Bunn et al., 2000), leading to
cumulative mortalities of 9899%. Consequently, small shifts in the
mortality rate lead to large changes in survivorship and associated
recruitment. There is a large range of potential predators responsible for egg and larval mortality including gelatinous organisms,
amphipods, mysids and euphausiids, carnivorous copepods,
chaetognaths and sh (e.g. Bailey and Houde, 1989; Bunn et al.,
2000). However the knowledge on which are the actual predators
in situ is limited, in particular with respect to invertebrates. In
this sense, changes in the predator community and/or in prey

Corresponding author. Tel.: +34 946015503.

E-mail address: [email protected] (A. Albaina).
http://dx.doi.org/10.1016/j.pocean.2014.12.002
0079-6611/ 2014 Elsevier Ltd. All rights reserved.

abundance have the potential to impact the recruitment success

(e.g. Koster and Mollmann, 2000; Lynam et al., 2005; Richardson
et al., 2011; Irigoien and de Roos, 2011).
There are a number of competitors for food of the adult anchovy
such as sardine, sprat, mackerel and horse mackerel that also predate on their ELS (Szeinfeld, 1991) and the consequences of intraguild predation have been discussed (Irigoien and de Roos, 2011).
However, predation by marine invertebrate has been largely
absent from sh recruitment models although it is known that
euphausiacea (Krautz et al., 2007) and chaetognatha (Terazaki,
2005) may be important causes of mortality of anchovy eggs. For
example, predation by euphausiacea accounted for 2427% of natural mortality in the Chilean anchoveta Engraulis ringens (Krautz
et al., 2007) and 4778% of the natural mortality on northern
anchovy (Engraulis mordax) eggs and yolk-sac larvae (Theilacker
et al., 1993).
The main reason underlying this lack of knowledge on the role of
ELS predation in sh recruitment success is that studying predation
in the eld has proven challenging because accurate identication of
partially ELS remains in predator stomachs is difcult (Bailey and
Houde, 1989; Heath, 1992). Fish eggs and larvae can be counted in
the guts of other sh, but only for a few minutes/hours after ingestion as they are digested quickly (e.g. Hunter and Kimbrell, 1980;

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Folkvord, 1993; Schooley et al., 2008). In the case of invertebrates,

accurate identication of sh eggs, larvae and juveniles in the stomach contents is usually much more difcult because invertebrates,
particularly the crustacean, macerate their prey (e.g. Theilacker
et al., 1993; Vestheim et al., 2005; Coston-Clements et al., 2009).
Accurate quantication is an additional problem and Baker et al.
(2014) have recently proposed to limit analyses to presence/
absence data. Beside this, identifying remains visually (if possible)
is a labour-intensive task, limiting the number of samples that can
be analyzed. However, a growing number of studies have employed
molecular methods to identify target species in the guts of predators
(e.g. reviews in Symondson, 2002 and King et al., 2008). Briey, two
molecular approaches have been applied intensively to detect predation in the eld: (1) immunoassays and (2) DNA based assays.
Although immunoassays have been used to measure predation on
other anchovy species ELS (Theilacker et al., 1986, 1993; Krautz
et al., 2003, 2007), DNA-based methods present several advantages
making them the method of choice when facing the molecular
determination of diet. Briey DNA-based methods are not only easier and cheaper to develop than immunoassays, but also allow rapid
screening against a multitude of different prey likely to be encountered in the eld (e.g. Symondson, 2002). To date all the methods
targeting prey DNA in predators have relied upon a polymerase
chain reaction (PCR) step. Among them, those involving monitoring
via real-time PCR detection of an species-specic DNA sequence are
faster, more sensitive (capable of detecting DNA traces) and offer
improved specicity over conventional PCR approaches generally
involving visualization of gel bands (e.g. McBeath et al., 2006). This
allows the reliable species assignment of a single sh egg DNA overcoming traditional methods limitation (e.g. Fox et al., 2005, 2008;
Minegishi et al., 2009). Examples of real-time PCR application to
characterize marine animals diet include copepods (Nejstgaard
et al., 2008; Troedsson et al., 2009; Durbin et al., 2008, 2011), decapods (Tbe et al., 2010; Albaina et al., 2010, 2012; Cleary et al., 2012;
Redd et al., 2014), shes (Hunter et al., 2012; Fox et al., 2012;
Baerwald et al., 2012) and mammals (Bowles et al., 2011).
The Bay of Biscay anchovy (Engraulis encrasicolus), a species
with large socio-economic impact in the Bay of Biscay area (e.g.
Uriarte et al., 1996), experienced, since 2002, a succession of low
recruitments, resulting in the collapse of the stock in 2005 that
led to successive closures of the shery until reopening in 2010
(Andrs and Prellezo, 2012; ICES, 2013). It has been suggested that
anchovy recruitment in the Bay of Biscay could be partially controlled by ELS predation (Irigoien et al., 2007) and in particular
by intraguild predation (Irigoien et al., 2007; Irigoien and de
Roos, 2011). However, predation ecology on ELS has hardly been
studied and only limited to sh predators (Goi et al., 2011;
Bachiller, 2013). In this study, our objectives were to determinate
the range of predators consuming anchovy ELS and to provide an
estimation on the contribution of mortality by predation in Bay
of Biscay anchovy eggs survival. To accomplish this, we designed,
validated and applied a molecular method capable of detecting
traces of E. encrasicolus DNA.

Materials and methods

European anchovy DNA detection assay
Assay design
Based on the high sensibility and specicity of the real-time PCR
based TaqMan assays and on the suitable characteristics of mtDNA
for trace DNA detection (see reviews in Symondson, 2002 and King
et al., 2008), we aimed to design and validate a TaqMan assay targeting a mtDNA specic sequence for European anchovy (Engraulis
encrasicolus) detection in predators stomach contents.

83

We screened 5 potential mtDNA target genes [12S rRNA, 16S

rRNA, cytochrome oxidase subunit I (COI), cytochrome-b (cytB)
and D-loop] using a set of E. encrasicolus mtDNA sequences [see
http://tomato.bio.trinity.edu/manuscripts/12-2/mer-11-0256.pdf,
accompanying paper of Molecular Ecology Resources Primer
Development Consortium et al. (2012)] along with other clupeids
sequences retrieved from NCBIs GenBank database (http://
www.ncbi.nlm.nih.gov/genbank/). We selected cytB as target due
to the absence of insertions and deletions, the combination of conserved and variable regions along it and, to the relatively larger
availability of clupeid specic sequences in public databases. In
order to include the highest E. encrasicolus genetic variability in
our assay, further sequencing of the cytB gene was performed following Jrme et al. (2008) primers (CytBI-7F and TruccytB-R) and
PCR conditions. Combining these new sequences (GenBank accession numbers KF972469KF972488 and KF972503KF972539)
with all the GenBank available matching sequences (9th May
2012 search) a total of 81 E. encrasicolus individuals sequences, distributed along the species distributional area, were then aligned
using the Clustal W algorithm (Thompson et al., 1997) implemented in Bioedit (Hall, 1999) (Table 1; Supplementary Material
Table 1). A total of 482 bp were considered for the assay design
including the in silico assessment of both intra- and inter-species
specicity. To accomplish this, sequences from other 12 clupeid
species, aiming to cover the highest possible spatial coverage
within them, were either produced within this study (sequenced
with the previous primers and conditions; accession numbers
KF972489KF972502 and KF972540KF972543) or retrieved from
GenBank (9th May 2012 search) adding a total of 105 sequences
that were then aligned against the target species (Table 1, Supp.
Mat. Table 1). Providing that the assay differentiate E. encrasicolus
from the closest genetically relatives, the assay will then discriminate from the less related ones. From these 186 cytB aligned
sequences, a real-time PCR based TaqMan assay for E. encrasicolus
DNA discrimination was designed using Primer Express software
(Applied Biosystem) and manufacturer recommendations. Both
primers and probe were located on conserved areas for the target
species but showing variation against the rest of species (Fig. 1).
The assay amplied a total of 87 bp. BLASTn algorithm (https://
blast.ncbi.nlm.nih.gov) was used to discard the alignment of other
taxa apart from clupeids. The 50 end of the MGB (minor groove
binding) TaqMan probe was labelled with uorescent dye FAM
(Applied Biosystems) and the 30 end labelled with a non-uorescent quencher. The 15 bp E. encrasicolus probe showed at least 1
SNP against the rest of clupeid species except for E. japonicus where
only a SNP in the last position of the reverse primer was noted. This
does not compromise the objectives of the research as these species never co-occur in the wild.
Assay validation
The assay intra- and inter-species specicities were tested
against DNA extracted from wild specimens distinct from the ones
used in the assay development. Assay sensibility was also tested
against a series of E. encrasicolus DNA dilutions including both muscle tissue and single eggs. DNA extractions and real-time PCR settings for assay validation were based in those established by
Albaina et al. (2010, 2012). DNA was extracted using a modied salt
extraction protocol (Aljanabi and Martinez, 1997). Following dissection, tissues were partially homogenized in 1.5 ml autoclaved
Eppendorf tubes containing 675 ml of extraction buffer (30 mM
Tris, 10 mM EDTA and 1% SDS) and left overnight at 55 C to digest
after addition of 10 ll Proteinase K (20 mg/ml; Roche). Dissection
tools were amed with ethanol after each sample and an extraction
blank (EB) control (negative control, where no tissue is added to the
extraction buffer prior to DNA extraction protocol) was included
every 11 samples to prevent cross-contamination. Next, 225 ll

84

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Table 1
Species and geographical origin of the 186 cytochrome-b (cytB) sequences used in the study. NEA stands for North-Eastern Atlantic samples (Celtic Sea, North Sea and Baltic Sea);
Bay of Biscay includes Cantabrian Sea samples.
Species

Code

NEA

Bay of Biscay

W. Medit.

E. Medit.

Canary I.

South Africa

Unknown loc.

Engraulis encrasicolus
Sprattus sprattus
Clupea harengus
Sardina pilchardus
Sardinella aurita
Sardinops sagax
Alosa fallax
Alosa alosa

EENC
SSPR
CHAR
SPIL
SAUR
SSAG
AFAL
AALO

81
23
14
10
9
3
9
3

21
4
6
2

27
17
2
3

2
2

1
3

2
2
1
2

Other anchovy species

Engraulis japonicus
Engraulis anchoita
Engraulis ringens
Engraulis mordax
Anchoa mitchilli

EJAP
EANC
ERIN
EMOR
AMIT

n
19
5
5
3
2

Total

2
2

5
2
1

2
7

Others

Origin
Japon
Brasil
Peru
USA (Pacic coast)
USA (Gulf of Mexico)

186

Fig. 1. Engraulis encrasicolus DNA assay design. Alignment showing the location of the TaqMan assay primers and probe on the cytochrome-b gene. Forward primer (50 30 ):
TTCTTACATGAATCGGAGGTATGC, Probe (50 30 ): CGAACACCCATTCAT, and Reverse primer (50 30 ) GGAARATAGAGAAGTAGAGTAGCGATGCT (reverse complement of the
sequence shown in the alignment, grey highlighted). Species codes as in Table 1.

5 M NaCl was added to the homogenate and briey vortexed. After

centrifuging for 5 min at 13,000 rpm, 450 ll of the supernatant was
collected in new 1.5 ml Eppendorf tube, 900 ll of 100% ice cold ethanol added before leaving for 1 h at 80 C. After 30 min centrifugation at 13,000 rpm, the liquid phase was removed and 1 ml of 70%
ethanol was added to wash the remaining pellet. Finally, after
10 min centrifugation at 13,000 rpm the ethanol was poured off,
the DNA pellet was then dried at 37 C, then resuspended in
100 ll ultrapure H2O and stored at 20 C. TaqMan assays were
run on an Applied Biosystems 7900 real-time sequence detection
system. 10 ll volume reactions were run in 384-well reaction
plates using Optical Adhesive Covers (Applied Biosystems). Each
reaction contained 0.083 ll of 60X assay (corresponding to 125
nM of anchovy probe and 450 nM of both the F and R primers),
5 ll of Brilliant III Ultra-Fast QPCR Master Mix (Agilent Technologies), 0.15 ll of ROX reference dye (1 mM; Agilent Technologies),
1.25 ll BSA (#B9001S New England Biolabs; 10 mg/ml), 2.517 ll
of ultrapure H2O and 1 ll extracted DNA. DNA yield (ng ll1) and

purity indexes [determined from the absorbance (A) at different

wave lengths (nm); A260/A280 and A260/ A230 ratios] for the
extractions were determined using a ND-1000 Spectrophotometer
(NanoDrop). Plates were run under FAST cycling conditions on a
single dye layer with 20 no template controls (NTCs) and 12 positive controls (DNA extracted from anchovy muscle tissue) per
384-well plate. The assay was run using 40 PCR cycles following
Master Mix manufacturers conditions: after a rst stage of 3 min
at 95 C, the run comprised 40 cycles of 5 s at 95 C followed by
15 s at 60 C. After PCR, the results were analyzed using the
Sequence Detection Software version 2.3 (Applied Biosystems)
and threshold cycle (Ct) values were computed. Ct values are
directly correlated with the number of copies of target DNA present
in the reaction. A lower Ct value means higher amount of target
DNA in the reaction and, vice versa. All amplication curves were
checked visually to remove false positive/negative signals.
The designed real-time PCR based TaqMan assay amplied
100% of E. encrasicolus samples from different locations (including

85

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

>10 individuals sampled at Bay of Biscay and Celtic Sea waters).

Ten-fold dilutions of an E. encrasicolus DNA extract (standard
curve) were used to estimate the detection limit of the assay (sensibility). Fig. 2a shows the magnitude of the detection signal
(expressed as Ct value) along the decreasing values of template
DNA (ng). The TaqMan assay was able to detect up to 0.0005 ng
of anchovy DNA. However, full (100%) reproducibility among replicas was only attained at amounts of target DNA P 0.005 ng. To
assure reproducibility, the assay detection limit for a reliable
detection of anchovy DNA was set at 35.4 Ct value (taking into
account both average Ct and SD values of the last dilution yielding
a Ct value for all the replicas). The efciency of the reaction was
>90% (following the equation in Galluzzi et al., 2004) for the different dilution series. As expected a 3 Ct units increase corresponded to each 10 fold dilution (e.g. Albaina et al., 2010;
Jungbluth et al., 2013). The high sensibility of the assay allowed
detecting DNA extracted from a single anchovy egg. When testing
20 E. encrasicolus eggs, including all development stages (Moser
and Ahlstrom, 1985), every individual egg gave a positive signal
well within the detection limits of the assay (average Ct of 26.6,
SD = 2.3). No relationship with stage development was reported
(Supp. Mat. Fig. 1). Roughly 0.5 ng of DNA was assayed per anchovy
egg. In this sense, the amount of DNA assayed from a single egg
corresponded to 100 times the assay detection limit (Fig. 2a).
Regarding inter-species specicity, the assay was tested against
DNA from clupeid species potentially cohabiting with anchovy in
the Bay of Biscay or surrounding regions. This included testing
specimens of Sardina pilchardus (n = 15), Sprattus sprattus (n = 10)
and Clupea harengus (n = 20) from the Bay of Biscay, Celtic sea

(a)

40

Ct (threshold cycle)

35

and Irish Sea. The assay tested negative for 100% of the cases for
sardines and herrings including a wide range of assayed DNA
amounts (up to >200 ng). However, non-complete specicity
appeared with sprat DNA extracts. Six out of ten sprats gave signals
below the 35.4 threshold. For these six cases an average Ct of 34.2
corresponded to an average assayed DNA amount of 150 ng. Conversely, 150 ng of anchovy DNA would correspond to Cts of
17.3. Furthermore, none of the 10-fold dilutions of sprat DNA
yielded a positive signal (Fig. 2b). If we assume the same DNA
amount per egg as in anchovy, 300 sprat eggs would be needed
to yield a 34.2 signal that is around the Ct value corresponding
to 1/100 dilution of a single anchovy egg (Fig. 2a).
This contrasting sensibility of the assay granted the applicability of the assay to measure anchovy predation in the eld in a reliable way. In this sense, new Ct thresholds were dened assuming
that the entire DNA extracted from the stomach contents corresponded to raw (undigested) sprat DNA (worst case scenario).
While the 35.4 threshold was sufcient when assaying <50 ng of
DNA extracted from stomach contents, a threshold of 32.4 Ct units
corresponded to assays with between 50 and 500 ng DNA and,
nally, a 29.4 Ct units threshold did to 5005000 ng DNA ones
(Fig. 2b). This prevented false positive signals arising from hypothetical sprat DNA presence. Finally, a further distinction had to
be done as, while in sh only stomach contents were extracted,
the whole animal was extracted in the relatively small size invertebrates comprising the zooplankton predators (see Section Detection of anchovy DNA within BIOMAN 2010 stomach contents). For
sh stomachs and large zooplankton, where a partial dissection
was performed, the entire amount of extracted DNA was

24; 0
34,7 (0,7)
24; 10

30
24; 0
30,9 (1,1)

25

24; 0
26,0 (0,28)

24; 0
22,4 (0,21)

20

24; 0
18,9 (0,26)

15
3; 0
15,9 (0.23)

10
5
0
0.0001

0.001

0.01

0.001

0.01

0.1

10

100

1000

0.1

10

100

1000

(b) 40
Ct (threshold cycle)

35
30
25
20
15
10
5
0
0.0001

Assayed DNA (ng)

Fig. 2. E. encrasicolus DNA assay validation. (a) Standard curve: rhombus representing threshold cycle (Ct, left axis) values plotted against serial dilutions of E. encrasicolus
template DNA (ng, bottom axis); assay detection limit (0.005 ng DNA for a Ct of 35.4), average Ct values and standard deviations (in brackets) are superimposed along with
the number of replicas tested and the number of assays giving a negative signal. Grey triangles signal one E. encrasicolus egg results (from right to left: undiluted, 1/10 and 1/
100 dilutions from one egg extract). (b) Superimposed lines and squares showing the thresholds dened to call a positive assay [Ct thresholds as a function of the amount (ng)
of assayed DNA]; circles represent the results for all the tested Sprattus sprattus raw tissues DNA extractions. See Section Assay validation for further information.

86

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

considered as potentially raw sprat tissue DNA. However, for the

remaining zooplankton individuals extracted as a whole (1096
out of 1174), we assumed a 25% of the whole individual extracted
DNA corresponding to stomach contents DNA although this is also
an overestimation reinforcing the conservative nature of the assay
thresholds.
May 2010 sampling
The BIOMAN surveys are carried out annually by AZTI-Tecnalia
since 1987 in the Bay of Biscay during the E. encrasicolus spawning
peak (MayJune; e.g. Motos et al., 1996; Santos et al., 2011) and are
designed to cover the whole spawning area of this species. These
surveys main objective is to determinate the Bay of Biscay
anchovy spawning stock biomass (SSB) following the DEPM [Daily
Egg Production Method; see Lasker (1985) and Somarakis et al.
(2004) for further information] as to provide sheries-independent
information for stock assessment. The 2010 BIOMAN survey was
carried out from the 5th to the 20th of May and involved two
research vessels (RV), namely RV Investigador and RV Emma
Bardn for, respectively, plankton and sh sampling. For a more
detailed description of 2010 BIOMAN survey, see Santos et al.
(2011).
Anchovy eggs sampling
A grid of 484 stations was sampled using vertical hauls of a
150 lm PairoVET net tted with a owmeter and lowered to
100 m or 5 m above the bottom at shallower stations. After allowing 10 s at the maximum depth for stabilization, the net was
retrieved to the surface at a speed of 1 m s1. Consecutive stations
were 3 nautical miles apart located in transects spaced 15 nautical
miles apart covering the Bay of Biscay from 43.3N to 46.9N and

from 1.2W to 4.5W (Supp. Mat. Fig. 2). PairoVET samples were
preserved immediately after collection in formaldehyde 4% buffered with sodium tetraborate in sea water. After six hours of xing, anchovy, sardine and other eggs species (other sh eggs
category) were identied, sorted out and counted. Apart from
PairoVET samples, the Continuous Underway Fish Egg Sampler
(CUFES, Checkley et al., 1997) was used to record the eggs found
at 3 m depth with a net mesh size of 350 lm.
Adult sh sampling
Sardine (S. pilchardus) and sprat (S. sprattus) samples were
obtained by pelagic trawling on board R/V Emma Bardn coinciding in space and time with the plankton sampling (see Santos
et al., 2011 for more information). Although the shing objectives
are stock evaluation of anchovies, other sh species are captured.
Immediately after shing, hauls where sardine and/or sprat were
present in sufcient number (aiming to sample 30 individuals
per species and haul but with a minimum of 10 to proceed) were
selected for sh sorting (Fig. 3). In this sense, it has been reported
that 20 stomachs per sh species ensured the detection of 75% of
the ingested prey groups in the Bay of Biscay (Bachiller, 2012). Less
than one hour, after the haul was retrieved, was needed to complete sorting. Sardine and sprat whole individuals were preserved
at 20 C. Back in the laboratory, with the sh still frozen, individual sardines and sprats were weighted, measured, sexed, gonad
stage determined following ICES methodology (ICES, 2008) and
the stomach (cardiac stomach and pyloric caeca) was dissected
and preserved in 50 ml Falcon tubes lled with fresh 100% ethanol
for further treatment. Dissecting tools were ame sterilized after
each dissection. An insufcient number of individuals in certain
hauls prevented statistical comparison among hauls; nevertheless,
statistical tests were applied to both species considering all the

Fig. 3. Prey and predators spatial location. E. encrasicolus egg abundance (ind. 1000 m3, scale superimposed) during the BIOMAN 2010 campaign obtained with SURFER 10
(Golden Software) with PairoVET data. Apart from this, while the grid of black dots signals PairoVET stations (parallel cross-shelf transects perpendicular to the coast),
locations of the macrozooplankton stations (MIK hauls, red stars) and sh hauls (pelagic trawling; blue crosses) are also shown. A pink circle locates the MIK haul performed
to collect live animals for the digestion experiment (see Section Digestion experiment). Isobaths of 100, 200 and 1000 m are shown (dotted lines) along with the position of
the Adour and Gironde river mouths and the Arcachon Bay. The two main spawning centers for anchovy, at the shelf in front of the Gironde river mouth (Gironde shelf one)
and at the shelf-break between Arcachon Bay and Adour river mouth (shelf-break one), can be easily identied.

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

tested animals together and also dividing them among three

dened temporal ranges [midday (11:3014:30 h local time), late
afternoon (18:3020:00 h) and midnight (22:3000:30 h); respectively, MIDD, LAFT and MIDN].
Macrozooplankton sampling
A total of 5 MIK (Methot Isaac Kidd) net samples, with a mesh
size of 1 mm and a mouth area of 1 m2, were collected in order
to characterize the macrozooplankton community and to sort
potential predators of anchovy eggs for assay testing (Fig. 3). All
but one haul were towed at areas of high anchovy eggs abundance
and during the night as to maximize predation events detection
(e.g. Kaartvedt et al., 2002; Mauchline, 1998; Jumars, 2007). With
the ship at 2 knots, oblique MIK net tows were carried out from
50 to 60 m (below the incipient thermocline) to the surface at a
speed of 3 m min1. Samples were preserved immediately after
collection in 100% ethanol. The ethanol was changed after 24 h
(onboard) and at least another time back in laboratory as to assure
optimum preservation of both morphology and DNA quality. The
qualitative and quantitative analysis of MIK net samples was carried out under a stereoscopic microscope and identication was
made to genus or species level when possible (Table 2). The
assayed taxa were selected based on their reported carnivorous
or, at least omnivorous, potential feeding character. In each sample, all the large animals were sorted, (mostly the myctophidae
Benthosema glaciale, the pteropod Cymbulia peroni, amphipods over
9 mm total length (TL), malacostracans over 7 mm cephalothorax
length (CL), and salps over 20 mm). Then, the remaining sample
was aliquoted using a Motoda plankton splitter (Motoda, 1959;
Van Guelpen et al., 1982) and aliquots were sorted until a minimum of 300 individuals (including all categories except gelatinous
organisms) were counted and measured. Further aliquots were
analyzed, if necessary, until at least 50 individuals from any of
the taxa selected for assay testing were sorted (Table 2). Potential
predators did not include gelatinous zooplankton and sh larvae
(except for Myctophid larvae than showed a better condition
related to their larger size) due to a relatively damaged condition
caused by an inappropriate sampling device. Individuals to be
assayed were transferred to 2 ml microtubes (Sarstedt) with fresh
ethanol until DNA extraction. Multivariate analyses of the sampled
stations and the relevant zooplankton taxa were carried out using
via DCA (detrended correspondence analysis) using version 4.5 of
CANOCO (ter Braak and Smilauer, 2002) applied to log10 transformed (x + 1) abundance values using only those taxa that contributed more than 5% of the zooplankton community abundance
at any of the stations when gelatinous organisms were not considered (Supp. Mat. Table 2).
Both PairoVET and MIK nets were tted with a RBR XR-420 CTD
(Conductivity, Temperature, and Depth proler; Sidmar) with a
uorescence sensor (Seapoint Chlorophyll Fluorometer; Seapoint
Sensors, Inc.). Water density (expressed as sigma-t) proles were
calculated. In addition, surface temperature and salinity were
recorded in each station with a manual thermosalinometer WTW
LF197.
Digestion experiment
The capability of detecting the DNA extracted from one anchovy
egg does not assure the assay capacity to detect predation events.
Firstly, how digestion time affects the ability of the probe to detect
the target DNA has to be estimated (e.g. Greenstone et al., 2013). To
estimate the digestion time effects we considered previously published results (Albaina et al., 2010; Hunter et al., 2012), combined
with an experiment for macrozooplankton performed onboard
during the campaign. An extra MIK haul was carried out (46N,
2.42W; Fig. 3) to collect a sufcient number of live macrozooplankton organisms for digestion experiments purposes. Although

87

a lower retrieving speed was used (11.5 m min1) combined with

a mere 016 m sampling depth, only Brachyuran larvae seemed
active and in good condition (based on swimming behavior and/
or response to stimulus) and thus were selected for the experiment. A total of 16 Liocarcinus depurator and 12 L. holsatus megalopae (identied following Ingle, 1991), were gently transferred in
similar numbers to four 2 L borosilicate glass bottles (Schott), containing 20 lm ltered sea water, mounted on a plankton wheel
that maintained sea surface temperature. After transferring animals, these were starved for 48 h before feeding trials commenced
to assure predator gut clearance and animals feeding shortly after
food is provided. As when the acclimation phase ended no anchovy
eggs were present in the area, raw anchovy tissue was used to feed
animals. Experimental animals were fed ad libitum for 2 h. Then,
six actively swimming individuals were preserved (end of the feeding period; t = 0 h) and the remaining animals were transferred
into new bottles with clean water and no anchovy tissue. Around
ve to nine actively swimming individuals were sampled at 3 different time points, 3 h, 6 h and 12 h after the end of feeding period.
Animals were preserved in 100% ethanol in 2 ml microtubes until
further DNA extraction. Half-life detectability rates of the target
DNA (T50 values; the time after which only half of the individuals
test positive for prey DNA) and maximum detection times were
estimated.
Detection of anchovy DNA within BIOMAN 2010 stomach contents
Back in the laboratory sardine and sprat stomach contents were
dissected for DNA extraction. At the same time, a visual fullness
index was recorded for both cardiac and pyloric caeca: an index
of 1 indicated stomachs with either not visible contents or with
less than 1/10 of the volume lled; 2 indicated between 1/10 and
2/3 of the volume lled and 3 indicated more than 2/3 of the stomach volume lled but not completely full; for the latter a category
of 4 was assigned. Prior to extraction, individual organisms were
placed over a highly absorbent wiper and washed with distilled
water using a Pasteur pipette. Forceps and dissecting tools were
ame sterilized after each animal. DNA was extracted in 1.5 ml
Eppendorf tubes following the salt extraction based method
detailed in Section Assay validation and including an extraction
blank control every 11 samples to detect cross-contamination.
For krill and other large malacostracans, a plastic pestle, treated
with bleach and UV radiation after each use, was used to aid
homogenization. After extraction, DNA was resuspended in
100 ll ultrapure H2O and stored at 20 C. DNA yield (ng ll1)
and DNA purity was assessed with the spectrophotometer. TaqMan
assays were run under FAST cycling conditions as in Section Assay
validation with 20 NTCs and 12 positive controls per 384-well
plate. In order to overcome PCR inhibition, apart from the addition
of 1.25 lg ll1 BSA to the reaction, two 10-fold dilutions were
tested for each sample with at least one of those corresponding
to a quantity tested of <50 ng DNA (stomach contents extract) as
this has been shown to reduce inhibition incidence to a minimum
(Albaina et al., 2010). The dened thresholds (Ct versus ng of DNA
tested; Fig. 2b) were applied to call a positive signal. Finally, the
percentage of positive signals per haul/station was computed.
Mortality estimation due to macrozooplankton predation
In situ data for both predator and prey abundances were available at MIK nets location (Table 3) allowing the mortality estimation at haul point. We made the conservative assumption that each
assay positive signal corresponded to one anchovy egg/larvae
eaten in the last 24 h. Our anchovy DNA detectability experiment,
limited to one developmental stage of 2 crab species, showed that
predation events were detectable during 3 h (Fig. 4); therefore,
an individual continuously feeding along the 24 h cycle could con-

88

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Table 2
Macrozooplankton species list, showing presence/absence (1/0) at each MIK net haul (A, B, C, D and E) along with average and maximum abundance (individuals 1000 m3).
While Assay column signals the taxa selected for the application of the real-time PCR assay for E. encrasicolus DNA detection, the Code column identies the taxa selected for DCA
analysis (see Section Macrozooplankton sampling).
A

Code

Gelatinous
Jellyshes
Siphonophora
Ctenophora
Salps
Gelatinous remains

1
1
1
1
1

0
1
1
1
1

1
1
0
0
0

1
1
0
0
1

1
1
0
0
0

JELLY
SIPHO

Non-Gelatinous
Tomopteris spp.
Cymbulia peroni
Diacria trispinosa
Pteropod spp.
Podon spp.
Calanus helgolandicus
Calanus robustior
Eucalanus crassus
Temora longicornis
Centropages typicus
Candacia armata
Euchirella rostrata
Euchirella curticauda
Metridia lucens
Pleuromamma robusta
Pleuromamma spp.
Euchaeta acuta
Euchaeta hebes
Euchaeta spp.
Paraeuchaeta gracilis
Paraeuchaeta norvegica
Paraeuchaeta tonsa
Paraeuchaeta spp.
Undeuchaeta major
Undeuchaeta plumosa
Undeuchaeta spp.
Acartia clausi
Oithona plumifera
Other/damaged Copepods
Conchoecissa imbricata
Conchoecilla daphnoides
Schistomysis ornata
Schistomysis/Paramysis spp.
Haplostylus normani
Haplostylus lobatus
Leptomysis gracilis
Mysid inmature/juvenile
Mysid damaged
Parathemisto abyssorum
Hyperia galba
Amphipod Gammaridae
Unknown amphipods
Diastylidae
Meganyctiphanes norvegica
Nematoscelis megalops
Thysanoessa longicauda
Nematobrachion boopis
Euphausiacea spp.
Pasiphaea sivado
Pasiphaea spp.
Systellaspis debilis
Solenocera larvae
Galatheidean megalopae
Paguridean megalopae
Zoea Porcellana
Zoea Galatheoidea Porcell.
Liocarcinus holsatus
Liocarcinus pusillus
Liocarcinus spp. damaged
Polybius henslowi
Discarded Swimming Crabs
Corystes cassivelaunus megalopa
Brachyuran zoeae
Other brachyuran megalopae

1
1
0
0
0
1
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0
0
0
1
1
1
0
0
0
0
0
0
0
1
0
0
1
0
1
1
1
0
1
1
1
1
0
0
0
1
0
0
0
0
0
0
0
1
1

1
1
1
1
0
1
1
0
0
1
0
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
1
1
1
0
0
1
0
0
0
0
0
0
0
1
1
0
1
0

0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
0
1
1
1
1
0
1
1
0
0
1
0
1
0
0
0
0
0
0
0
0
1
1
0
1
0
1
0
0
0
1
0
1
1

1
0
0
0
0
1
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
1
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
1
1

0
0
0
0
1
1
0
0
1
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
1
0
1
1
1
1
1
0
1
1
1
1
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
0
0
1
1
1

Assay

SALPS
GELAT

TOMOP

+
+
+

CHELG

EUCRO

PLERO

PARGR
PARTO

UNDPL

SCHOR
HAPNO

MYSID

+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+

+
+
+
+
+
+
+
+
+
+
+

MEGNO

+
+
+
+
+

EUPHA

SOLEN

+
+
+
+
+
+

+
+
+
+
BRAZO
BRAME

Average

Maximum

283.7
1280.6
1.6
401.2
494.4

1317.1
3855.9
5.4
1623.5
1864.3

49.0
2.1
1.2
0.3
2.9
439.3
2.5
2.0
11.4
5.0
30.6
100.0
8.7
5.8
45.2
5.4
12.1
32.0
24.1
56.5
9.6
63.2
2.1
34.9
66.4
1.2
4.7
2.9
14.6
2.1
2.1
351.9
37.6
390.1
34.3
4.8
460.6
26.4
0.1
0.2
21.8
3.1
51.1
351.2
16.0
2.1
1.2
181.9
3.2
4.2
0.1
335.5
27.5
11.4
72.3
2.9
26.3
2.9
2.9
0.2
1.3
51.4
472.8
200.8

229.0
5.4
6.2
1.6
14.3
1299.3
12.4
9.8
57.1
24.9
114.5
354.2
31.2
18.6
174.0
20.8
41.6
104.1
68.4
208.1
41.6
260.2
10.4
93.7
248.6
6.2
23.5
14.3
23.5
10.4
10.4
1716.8
188.1
1693.3
71.4
14.3
1716.8
117.6
0.7
0.9
85.7
14.3
141.1
1156.5
42.9
10.4
6.2
468.3
13.2
20.8
0.7
1367.3
114.2
57.1
258.7
14.3
128.5
14.3
14.3
0.8
5.9
257.0
1442.1
842.4

89

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Table 2 (continued)

Other decapod larvae

Chaetognatha
Echinodermata larvae
Oikopleura spp.
Ammodytidae
Merlangius merlangus
Benthosema glaciale
Myctophidae larvae
Clupeid larvae anchovy
Other sh larvae
Others (non-gelatinous)

1
1
0
0
0
0
1
1
1
1
1

1
1
0
0
0
0
1
1
1
1
1

1
0
1
1
1
0
0
0
1
1
1

1
1
1
0
0
0
0
0
1
1
1

1
1
1
1
0
1
0
0
1
1
1

Code

Np p  DC

where Np is the number of anchovy eggs consumed over the previous 24 h per unit area, p is the proportion of positive TaqMan assay
for a certain taxon, and DC is the estimated density of the predators
per unit area. Then, for each sampled location taking into account
every assayed taxon:

P
MP

DP

N
PP

NP

 100

Mortality estimation due to sh predation

The mortality exerted by sardine and sprat could be estimated
following the classical approach (Eqs. (3) and (4)) due to the existence of feeding duration and gut evacuation rates from the literature. In this sense, given the quantitative nature of the real-time
PCR technique (Ct M ng of anchovy DNA) and the amount of DNA
corresponding to one undigested/raw assayed egg (Fig. 2a) we
could estimate the number of anchovy eggs consumed by predator
from the obtained Ct value. This estimation represents the minimum number of eggs predated by each sh as the same Ct value
would correspond to a progressively higher number of eggs as
digestion time progresses. Apart from this, the availability of biomass estimations for Sardina pilchardus and Sprattus sprattus from
the parallel PELGAS 2010 acoustic survey (ICES, 2010b) along with
the E. encrasicolus daily egg production and mortality rates from
BIOMAN 2010 campaign (ICES, 2010a), allowed us to calculate
the (minimum) proportion of anchovy egg mortality that can be

Maximum

94.2
20.1
134.8
7.6
1.2
0.2
3.7
21.6
216.8
128.5
141.7

299.8
31.2
428.3
23.5
5.9
0.9
16.3
83.3
656.8
385.5
285.6

2461.5
4960.2
7421.6

4405.5
7897.4
10299.6

+
+
+
+
OCLUP

assigned to sardine and sprat predation. Regarding sardine, the

PELGAS 2010 biomass estimations for two discrete areas, the socalled Gironde and South-offshore ones, approximately corresponded to the two main anchovy spawning centers in 2010 (from
here onwards Gironde shelf and shelf-break ones; Fig. 3). For sprat,
however, the provided biomass included that of the sprats inhabiting Gironde shelf waters along with the more abundant ones at
northern locations thus preventing more accurate mortality estimation in the anchovy spawning areas (ICES, 2010b).
We calculated the proportion of anchovy egg mortality
explained by sardine and sprat predation (PC) following the
method given by Hunter and Kimbrell (1980) including the modication introduced by MacCall (1981) as in Szeinfeld (1991). We
estimate the number of eggs consumed daily per predator:

C EE  g  t0

where C is the average numbers of eggs eaten per kg of sh during

time t0 , EE average number of eggs per predator kg as estimated by
the real-time PCR assay, t0 is duration of feeding (h) and g is gastric
evacuation rate (h1). While whole day long feeding behavior of
sardine and one restricted to daylight hours in sprat (07:0021:00
for May in the Bay of Biscay) have been reported (respectively,
Garrido et al., 2007, 2008 and Peck et al., 2012), evacuation rates
of 0.2105 h1 (Nikolioudakis et al., 2011) corresponded to sardine
and were also applied to sprat due to the lack of comparable values
in the literature. Then, the proportion of anchovy egg mortality
caused by sh predation (PC):

where MP is the daily mortality at the sampling location exerted by

macrozooplankton predation and DP is the estimated abundance of
anchovy eggs per unit area.

Average

+
+
ECHIN

Gelatinous
Non-Gelatinous
Total

sume up to 8 times the amount detected in the last 3 h. However,

the variety of taxa involved and the lack of information about macrozooplankton feeding behavior and digestion times (e.g. Durbin
et al., 2011) made us consider the 1 positive assay = 1 egg/larvae
killed in the last 24 h as a reasonable conservative assumption
representing minimum estimation of the predation impact of macrozooplankton on anchovy. As an example, between 1.6 and 1.9
sh larvae (respectively, Brevoortia tyrannus and Anchoa mitchilli)
were ingested daily by the chaetognath Sagitta hispida in controlled
laboratory experiments where high encounter rates were favoured
(Coston-Clements et al., 2009). Due to the absence of anchovy larvae counts, daily mortality due to macrozooplankton predation
was then computed as the fraction of anchovy eggs eaten in the
last 24 h (Eqs. (1) and (2); adapted from Albaina et al., 2012). Both
prey and predator abundances were transformed to abundances
per unit area. Although the sampling depth were not the same
(Table 3) the vast majority of anchovy eggs are distributed within
the MIK haul sampling depth (Boyra et al., 2003; Coombs et al.,
2004). For each assayed taxon:

Assay

PC

ZC
C t =Gt

Z t 1  etZ

where Zc is the proportion of egg production consumed by sh predation (Ct/Gt), Ct is daily total anchovy egg consumption by sardine
or sprat (C predator biomass) and Gt is total daily anchovy egg
production; Zt is proportion of egg production lost due to all causes
of mortality (1  etz); Z is hourly instantaneous rate of egg mortality and t = 72 h (anchovy egg development time, following Motos,
1994).
Results
A total of 190 extraction blanks and 260 no template controls
(NTCs) were included in order to check cross-contamination. Only
2 extractions blanks and 2 NTCs gave a positive signal, which after
repetition in a different real-time PCR run showed negative signal
thus being more than likely associated to contamination due to
incorrect pipetting when setting the 384-well real-time PCR plates.
The same could explain the 2 out of 260 positive signals in the
NTCs included in the plates. Based on these results a marginal
0.9% of positive signals (4 out of 450) could be assigned to false
positives.

MIK-A

MIK-B

MIK-C

MIK-D

MIK-E

Date
Time of haul (local time)
Haul depth (m)
Bottom depth (m)

5/8/2010
03:28
65.6
1600

5/11/2010
04:46
63.6
1153

5/13/2010
01:12
45.8
79

5/14/2010
05:33
54.8
94

5/15/2010
01:13
53.5
73

Anchovy eggs (PairoVET)

Anchovy eggs at 3 m depth
(CUFES)

2291.9
22404.8

2568.4
28790.6

0.0
166.2

4075.2
25926.8

14482.1
60850.6

% + assays n assayed

Abundance % + assays n assayed

Abundance % + assays n assayed

Abundance % + assays n assayed

Abundance % + assays n assayed

Abundance

4.5
0.0

229.0
5.2
0.0
114.5
145.7
31.2
10.4
52.0
20.8
41.6
104.1
52.0
208.1
41.6
260.2
10.4
93.7
83.3
0.0
10.4
10.4
0.0
0.0
0.0
0.0
0.0
0.0
0.7
0.0
0.0
0.0
1156.5
42.9
10.4
0.0
2.6
20.8
0.7
0.0
0.0
0.0
0.0
0.0

6.2
5.4
6.2
0.0
354.2
12.4
18.6
174.0
6.2
18.6
55.9
68.4
74.6
6.2
55.9
0.0
80.8
248.6
6.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
599.7
37.3
0.0
6.2
13.2
0.0
0.0
6.2
0.0
0.0
0.0
0.0

0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1716.8
188.1
1693.3
70.6
0.0
1716.8
0.0
0.0
23.5
141.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
47.0
23.5
0.0
2.9
0.0

9.8
0.0
0.0
9.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
29.5
9.8
29.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
1367.3
0.0
0.0
0.0
0.0

0.0
0.0
0.0
28.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
42.8
0.0
257.0
71.4
14.3
556.8
0.0
0.9
85.7
114.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
257.0
114.2
57.1
128.5
14.3

12.5
0.0
0.0
0.0
0.0
0.0
0.0
10.0
0.0
15.0
0.0
4.2
0.0
0.0
11.1
0.0
0.0

0.0

3.8
0.0
0.0
0.0
0.0
100.0

22
8
0
8
13
3
1
4
1
4
10
4
20
4
24
1
8
9
0
1
1
0
0
0
0
0
0
1
0
0
0
52
6
1
0
3
2
1
0
0
0
0
0

0.0
0.0
0.0
0.0
50.0
0.0
0.0
0.0
0.0
0.0
12.5
0.0
0.0
10.0
8.3
0.0

0.0
0.0
0.0
10.0

0.0

1
7
1
0
14
2
1
15
0
1
6
10
8
1
8
0
10
24
1
0
0
0
0
0
0
0
0
0
0
0
0
62
6
0
1
10
0
0
1
0
0
0
0

0.0
25.0
5.6
0.0
2.7

0.0
66.7

50.0
100.0
0.0

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
73
8
72
3
0
73
0
0
1
6
0
0
0
0
0
0
0
2
1
0
2
0

100.0

0.0

0.0
100.0
0.0

72.7

1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
1
3
0
0
0
0
0
0
0
0
0
0
0
132
0
0
0
0

100.0

100.0
100.0
100.0
100.0
100.0
100.0
83.3
100.0

100.0
100.0
100.0
100.0
100.0

0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
18
5
1
40
0
1
6
8
0
0
0
0
0
0
0
14
9
4
9
1

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Tomopteris spp.
Cymbulia peroni
Diacria trispinosa
Candacia armata
Euchirella rostrata
Euchirella curticauda
Metridia lucens
Pleuromamma robusta
Pleuromamma spp.
Euchaeta acuta
Euchaeta hebes
Euchaeta spp.
Paraeuchaeta gracilis
Paraeuchaeta norvegica
Paraeuchaeta tonsa
Paraeuchaeta spp.
Undeuchaeta major
Undeuchaeta plumosa
Undeuchaeta spp.
Conchoecissa imbricata
Conchoecilla daphnoides
Schistomysis ornata
Schistomysis/Paramysis spp.
Haplostylus normani
Haplostylus lobatus
Leptomysis gracilis
Mysid inmature/juvenile
Parathemisto abyssorum
Hyperia galba
Amphipod Gammaridae
Diastylidae
Meganyctiphanes norvegica
Nematoscelis megalops
Thysanoessa longicauda
Nematobrachion boopis
Pasiphaea sivado
Pasiphaea spp.
Systellaspis debilis
Solenocera larvae
Galatheidean megalopae
Paguridean megalopae
Liocarcinus holsatus
Liocarcinus pusillus

90

Table 3
Results of the application of the assay targeting Engraulis encrasicolus DNA to the selected macrozooplankton predators. MIK hauls data including E. encrasicolus eggs abundance (ind. 1000 m3) are shown along with the percentage of
the assays testing positive for E. encrasicolus DNA. The total number of predators assayed per species and station is also shown as well as the abundance of these putative predators at the sampled point (ind. 1000 m3).

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

91

3171.5
68.6
5.6
1952.9
220
3.6

0.0
25.0

231
Total

0.0
0.0

4.8

2948.4

0.0

0.0
18.6
31.1
0.0
0.0
16.3
24.9
0
0
4
0
0
21
4
0.0

156
5723.6
248

1613.2

99.5

100.0

66.7
50.0
0.0
0.0
0.0
0.0
0.0

10.4
62.4
31.2
0.0
0.0
2.0
83.3
1
3
3
0
0
3
9
0.0
0.0
0.0

1
0

0.8
0.0
MIK-B

0
0

100.0
MIK-A

Polybius henslowi
Corystes cassivelaunus
megalopa
Other brachyuran megalopae
Other decapod larvae
Chaetognatha
Ammodytidae
Merlangius merlangus
Benthosema glaciale
Myctophidae larvae

Table 3 (continued)

214

842.4
299.8
28.6
0.0
0.9
0.0
0.0
59
13
2
0
1
0
0
100.0
100.0
100.0

23.5
70.6
0.0
5.9
0.0
0.0
0.0
1
2
0
4
0
0
0

12
2
1
0
0
0
0

0
0
0
0

127.9
19.7
9.8
0.0
0.0
0.0
0.0

0
18

MIK-D

0.0
0.0
MIK-C

0.0
0.0

100.0

MIK-E

0.0
257.0

Detectability of anchovy DNA over predators digestion

The E. encrasicolus DNA detectability along digestion time was
tested with an experiment during the plankton sampling cruise
(Section Digestion experiment). Fig. 4 shows the decay of E. encrasicolus DNA detection rates in two Brachyuran species along the
digestion process (up to 12 h after ingestion) at temperatures ranging from 13.2 to 14.4 C. For both species (namely Liocarcinus depurator and L. holsatus) the same signal decay pattern was reported
showing comparable detectability along digestive process. Half-life
detection rates (T50) of 5 h were reported for E. encrasicolus DNA
within stomach contents. Positive signals could be measured up to
6 h after ingestion (2 out of 9 individuals) but not after 12 h.
Plankton survey
Anchovy eggs distribution
Anchovy eggs were mostly distributed in two main areas that
correspond well with the two main spawning centers for the species in the last decade (Bellier et al., 2007; ICES, 2010a). The rst
center was located in front of the Gironde river mouth in the inner
shelf (<100 m) waters (from here onwards Gironde shelf one)
and the other in the shelf-break region between Arcachon bay
and Adour river locations (shelf-break one; Fig. 3). Abundances
up to 19.8 eggs m3 were recorded. All but one (MIK-C) of the ve
MIK hauls collected with predation assessment purposes were carried out in E. encrasicolus eggs high density areas (Fig. 3, Table 3)
with a maximum of 14.5 eggs m3 (MIK-E). Fish hauls were performed in areas with a moderate to high anchovy eggs abundance
(up to 5 eggs m3 in two of the hauls and never below
0.1 eggs m3; Table 4).
Macrozooplankton abundance and distribution
A total of 80 taxa were identied from MIK samples and 52 of
them were sorted for assay testing (Table 2). The most abundant
taxa apart from gelatinous organisms (mainly siphonophores and
salps, 34% of total abundance), were the euphausiid Meganyctiphanes norvegica, mysids (mainly Schistomysis ornata and Haplostylus normani), decapod larvae (Brachyuran and Solenocera larvae
being the most abundant) and the copepod Calanus helgolandicus.
The copepods were the most diverse group with 19 species identied. Interestingly, no anchovy egg or larvae was recorded. When
plotting the most abundant taxa (21 taxa; Table 2) in a detrended
correspondence analysis (DCA) with CANOCO software (see Section Macrozooplankton sampling), two assemblages were distinguished (Fig. 5; codes as in Table 2). Stations collected at the
shelf-break (MIK-A and B) occupied the left part of the biplot and
were characterized by salps, euphausiids and large copepod species. Gironde shelf stations (MIK-C, D and E) occupied the opposite
section of the DCA plot and were characterised by decapod larvae
and mysidacea. Macrozooplankton abundances over 7 ind. m3
were reported in the latter stations (with a maximum of
10.3 ind. m3 in station E) whereas in oceanic samples values were,
respectively, 5.5 and 6.3 ind. m3 for stations A and B (Supp. Mat.
Table 2).
Detection of anchovy DNA within macrozooplankton stomach
contents
From 1069 organisms tested (ranging from 156 to 248 individuals assayed per MIK haul), 353 yielded a positive signal including
32 different taxa (Table 3). While shelf-break stations presented
low frequencies of positive signals (4.8% and 3.6% of the individuals
testing positive for anchovy DNA for, respectively, MIK-A and B),
the Gironde shelf ones presented higher frequencies ranging from
5.6% in Station MIK-C to 68.6% in MIK-D and to 99.5% in station
MIK-E. Interestingly, predation incidence within the Gironde shelf

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

% testing positive

92
5/6
2/2

100
90
80
70
60
50
40
30
20
10
0

2/2
3/4

1/5
1/4
0/3
0/2

12

Digestion time (h)

Fig. 4. Detectability curves for Liocarcinus spp. onboard digestion experiment.
Respectively, white circles for Liocarcinus depurator specimens (n = 16; CW average
of 1.53 mm, 0.10 SD) and, white squares for Liocarcinus holsatus (n = 12; average CW
1.58 mm, 0.14 SD). Detectability half-life (T50) line is superimposed (see Section Digestion experiment for further information).

stations increased with prey abundance in a logarithmic way

(Fig. 6).
Sorted taxa in shelf-break stations were mainly represented by
large calanoid copepods and euphausiacea species. Among those,
the most abundant taxa, the euphausiid M. norvegica, hardly predated on anchovy eggs/larvae (only 2 individuals, out of 114, presented anchovy DNA remains within their contents). Regarding
copepods, a total of 7 species (Candacia armata, Euchirella curticauda, Euchaeta hebes, Paraeuchaeta gracilis, P. tonsa, Undeuchaeta
major and U. plumosa) presented positive assays (12 positive signals out of 215 copepod individuals for MIK-A and B). While one
out of 13 myctophidae larvae showed a positive signal, larger Benthosema glaciale did not test positive. Apart from these species,
positive assays were recorded also in the pelagic Polychaeta
Tomopteris spp. and the decapods Systellaspis debilis, Polybius henslowi and Pasiphaea sivado in the shelf-break stations (one positive
assay per taxa).
Regarding the Gironde shelf stations (MIK-C, D and E), mysid
species along with a variety of decapods larvae taxa represented
the bulk of organisms assayed. Only one targeted copepod species
(Candacia armata) was present in these samples. Mysids and the
cumacean Diastylidae spp. comprised the bulk of positive signals
in MIK-C. Regarding MIK-D and E, the bulk of positive signals corresponded to Solenocera larvae (especially in MIK-D where this
taxon comprised 85% of the assayed taxa) and Brachyuran megalopae, followed by the mysids assemblage (Schistomysis ornata,
Haplostylus normani, Haplostylus lobatus, Leptomysis gracilis and
unidentied or immature individuals) and other decapods larvae.
Apart from those taxa, positive signals in MIK-D and E hauls were
also recorded for Tomopteris spp., Candacia armata, the amphipods
Hyperia galba and Gammarids spp., Diastylidae spp., Chaetognatha
spp. and one juvenile Merlangus merlangus.
Adult sh sampling
Sardine and sprat hauls characterization
A total of 14 pelagic trawling hauls were collected in the
anchovy spawning area (Fig. 3). A total of 237 sardines and 213
sprat individuals were sorted and tested for anchovy DNA presence
in their guts (Table 4). While sardine was distributed along the
whole shed area including both anchovy spawning centers, sprat
was only caught in the Gironde shelf area. Sorted animals length
ranged from 138 to 227 mm for sardines and from 79 to 134 mm
in sprats with average weight of, respectively, 58 and 10 g (Supp.
Mat. Tables 3 and 4). Pyloric fullness was constantly high, showing
average values close to 3 in both species, and no signicant difference between the three different times of the day (Kruskal Wallis

Fig. 5. Macrozooplankton community. Detrended correspondence analysis (DCA;

CANOCO software) of BIOMAN 2010 macrozooplankton taxa. Species codes as in
Table 2, sampled MIK stations (MIK-A, B, C, D, and E) represented by black dots. The
cumulative explained variance by the rst axis was 81%; when adding the second
axis this shifted to 83%.

P > 0.001). Cardiac fullness index was more variable and there were
signicant differences between sampling times for both species
(Kruskal Wallis P < 0.001). Gut fullness data suggests feeding intensity peaking at dusk in both species but a higher preference for
dark hours in sardine than in sprat. Finally, no differences in both
fullness indexes were recorded regarding size and gonad stage.
Detection of anchovy DNA within sh stomach contents
Sardine samples. A total of 82% of the sardines stomachs showed a
positive signal (Supp. Mat. Table 3). Interestingly we found sardines with anchovy DNA remains in their stomachs in every sorted
haul (nine hauls with values ranging from 48% to 100% of positive
signals; Table 4). No signicant differences (Kruskal-Wallis tests)
were found when comparing predation incidence against dened
categories for cardiac and pyloric fullness (respectively, P = 0.06
and P = 0.21), size (P = 0.58), weight (P = 0.49) and female gonad
stages (P = 0.36). The same lack of signicance was reported when
comparing the three distinct times of day sampled (P = 0.14). Estimated anchovy eggs abundance at haul location and percentage of
positive signals are plotted in Fig. 7a. Although positive signals
incidence is higher at higher prey values, the correlation is not signicant (Pearsons rho 0.44, P = 0.24) and high percentages of sh
testing positive for anchovy DNA are found even at areas with relatively low abundances of anchovy eggs. The same lack of correlation corresponded to the number of anchovy eggs consumed per
predator, as estimated by Ct values, showing a slight increase associated with higher prey abundances (Pearsons rho 0.351,
P = 0.355; Fig. 7b). The maximum number of eggs estimated for a
sardine stomach was 252.
Sprat samples. A total of 42% of the assayed stomach contents for
sprat showed a positive signal for E. encrasicolus DNA presence
(Supp. Mat. Table 4). This value ranged between 10% and 63%
depending on the haul (Table 4). No signicant differences (Kruskal-Wallis tests) were found when comparing predation incidence
against cardiac and pyloric fullness indexes categories (respectively, P = 0.21 and P = 0.74) and the three distinct times of day
sampled (P = 0.06). Although not signicant the latter low P-values

93

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

100

90

90

80

Predation incidence
Gironde shelf
Shelf-break
MP
Gironde shelf
Shelf-break

70
60
50
40

80
70
60
50
40

30

30

20

20

10

10

0
0

200000

400000

600000

800000

1000000

MP (empty symbols)

% testing positive (full symbols)

100

0
1200000

Anchovy eggs (ind. 1000 m-2)

Fig. 6. Macrozooplankton predation on anchovy eggs. Full symbols represent the relationship between the occurrence of E. encrasicolus DNA remains, predation incidence, in
macrozooplankton taxa (rhombus and circles for the, respectively, Gironde shelf and shelf-break communities as clustered in CANOCO analysis) and the abundance of E.
encrasicolus eggs at the MIK haul location. The relationship between egg abundance and the daily mortality (MP; see Section Mortality estimation due to macrozooplankton
predation) due to macrozooplankton is shown by the empty symbols. For MIK-C station, due to the low resolution of PairoVET net counts at low egg abundances, anchovy egg
abundance was obtained by transforming eggs counts at 3 m depth (CUFES sampling device) to those at the water column (0.15 abundances at 3 m).

corresponded to a higher predation incidence in daylight hours

(Supp. Mat. Table 4). Regarding size and weight categories, there
was a signicant difference with larger sh presenting higher percentage of positives (P = 0.004 for size and P = 0.03 for weight).
However the average number of anchovy eggs consumed per individual, as estimated by assay Ct values, was similar in both size
categories (5.3 and 5 for, respectively, sh below or over 100 mm
total length). A maximum of 71 eggs was estimated for one sprat
individual. Finally, when plotting predation incidence and anchovy
egg abundance, the obtained pattern resembled the sardine one,
but with considerable lower incidence of positive signals given a
particular prey abundance and a higher dispersion of values (Pearsons rho 0.27, P = 0.56) (Fig. 7a). However, for sprat, the average
number of anchovy eggs consumed per individual decreased with
prey abundance (Pearsons rho 0.362, P = 0.425; Fig. 7b).

Predation mortality of anchovy eggs

Mortality due to macrozooplankton (MP)
Mortality due to macrozooplankton predation (MP, see Section Mortality estimation due to macrozooplankton predation)
ranged from 1% to 89% of the egg abundance (Fig. 6). Lowest mortalities were found at the shelf-break area (3.6% and 1.3% for,
respectively, MIK-A and MIK-B), whereas the Gironde shelf area
presented higher mortalities (from 14.3% and 14.6% at, respectively, MIK-D and E, to 89% at MIK-C).

Mortality due to sh (PC)

A minimum of 7.1% of the whole Bay of Biscay anchovy egg
daily mortality was explained by sardine predation when considering both anchovy main spawning centers (Table 5; see Section Mortality estimation due to sh predation for further
explanation). While a PC of 1.8% corresponded to sardine predation
at the shelf-break spawning center, the remaining 5.3% did to those
at Gironde shelf one. Regarding sprat, it has to be noted that this
species impact on anchovy egg survival is limited to the Gironde
shelf spawning center due to its distribution (ICES, 2010b). However, the lack of an acoustic biomass value related to the Gironde
shelf area from the PELGAS campaign prevented estimating mortality in an accurate way. If we consider the whole species biomass
for the Bay of Biscay when computing PC [an obvious overestimation due to the larger part of the 2010 sprat biomass being located
in coastal areas north of 47N (ICES, 2010b) well out of anchovy

spawning area], this would lead to a 7.2% of the anchovy egg daily
mortality.
Discussion
The method developed was able to detect up to 0.005 ng of
anchovy DNA (corresponding to 1/100 fraction of one egg signal)
in a reliable way (Fig. 2). Furthermore, the assay was able to detect
anchovy DNA up to 6 h after ingestion by Brachyuran larvae
(Fig. 4). This indicates that the assay is suitable to detect sh eggs
ingestion by macrozooplankton where target DNA concentrations
are expected to be low and the possibilities of microscopic analysis
limited. As the DNA based assay cannot distinguish between the
egg and larval stages we would refer to the predation on European
anchovy eggs/larvae throughout. However, when relating these
predation rates to prey elds and, when computing natural mortality by predation, we will restrict to anchovy egg distribution data
as these were the only available prey abundances. As anchovy eggs
would represent the bulk of anchovy early life stages (ELS) at the
sampling moment due to the species spawning dynamics (e.g.
Irigoien et al., 2007; Cotano et al., 2008) a signicant bias due to
the previous simplication is not to be expected.
Putative biases of the molecular method based diet assessment
First of all, we consider the impact of the assays lack of total
discrimination against sprat DNA (see Section Assay validation).
However, due to the reported different sensibility of the assay for
sprat and the target species DNA, this issue was overcome by setting new thresholds (Ct to extracted DNA ng threshold; Fig. 2b) to
call a positive signal in eld samples where, we assumed that the
entire DNA extracted from the guts was potentially undigested
sprat DNA. This prevented false positive signals coming from the
putative presence of sprat DNA within the contents. Apart from
this particular factor, two other factors, inherent to the application
of molecular methods to diet assessment, need to be evaluated:
cross-contamination and PCR inhibition. Given the sensitivity of
these methods, cross-contamination has to be assessed (e.g. King
et al., 2008). Special care has to be taken to prevent contamination
of samples from collection to the real-time PCR plate setting and,
with special care, in the dissection procedure. Sterility has to be
kept during all the process and both on-board and lab protocols
have to be adapted to these new issues including controls at different process points and dedicated laboratories for predator sorting,

94
Table 4
Fish hauls. Data on sh haul including local time, depth and captured sh. The information on assayed sardines and sprats is also shown (average values and range): cardiac and pyloric stomach visual fullness, female gonad stages,
length, weight and the percentage of sh testing positive for E. encrasicolus DNA presence along with the average number of consumed anchovy eggs (minimum value as estimated by Ct values, see Section Assay validation). For sprat no
female gonad data is provided as every individual was immature/resting or spent. The PairoVET based E. encrasicolus eggs abundance is also shown (ind. 1000 m3), including the temporal lag with the sh haul (hours earlier or later for
the sh haul with respect to the PairoVET net one). MIDD, LAFT and MIDN for, respectively, midday, late afternoon, and midnight.
1

10

11

12

13

14

Date
Time of haul (local
time)
MIDD-LAFT-MIDN
Haul depth (m)
Bottom depth (m)
Sardine Total (kg)
Sardine sorted for
assay (n)
Sprat total (kg)
Sprat sorted for
assay (n)
Other sh (kg)
Lag with DEPM
sampling

06-05-2010
18:56

07-05-2010
12:40

09-05-2010
19:12

10-05-2010
19:58

13-05-2010
12:19

14-05-2010
19:00

15-05-2010
14:29

16-05-2010
23:50

17-05-2010
11:30

18-05-2010
00:29

18-05-2010
12:10

18-05-2010
19:50

18-05-2010
22:41

19-05-2010
11:55

LAFT
17
>1000
7.1
30

MIDD
14
31
63.0
21

LAFT
70
100
4.1
30

LAFT
95
110
3.5
32

MIDD
88
105
5.6
29

LAFT
30
42
5.0

MIDD
35
48
4.8
32

MIDN
15
40
0.9
19

MIDD
113
125
2.3
31

MIDN
5
36
0.0

MIDD
40
60
5.0
13

LAFT
53
67
0.0

MIDN
5
38
0.0

MIDD
66
75
0.0

46
32

0.5
31

0.85
30

110
30

550
30

0.66
30

5.2
30

210.8
<36 h earlier

23.5
<36 h earlier

403.1
<12 h later

148.9
<36 h later

42.7
<3 days later

2000.0
<12 h later

26.5
<24 h earlier

0.0
<12 h earlier

144.5
<5 days later

31.0
386.5
<4 days later <3 days later

48.0
<3 days later

143.1
<3 days later

119.6
<3 days later

222.7

3249.9

3466.5

1458.1

2108.0

4943.6

105.4

1761.3

323.7

3492.1

1700.3

5677.5

3.1 (24)

3.3 (14)

2.3 (14)

1.1 (12)

2.0 (14)

2.5 (14)

1.1 (12)

1.1 (12)

3.2 (34)

2.8 (14)

2.7 (14)

3.0 (34)

2.9 (23)

2.7 (13)

1.9 (13)

1.0 (11)

3.5 (2.5)

3.3 (35)

3.5 (34)

3.5 (2.5)

3.9 (35)

3.9 (35)

4.5 (45)

3.7 (3.5)

3.4 (24)

2.4 (13)

1.6 (13)

2.9 (23)

3.0 (33)

Anchovy eggs
2694.6
(PairoVET)
Assayed sardine characteristics:
Sardine gut fullness 2.5 (14)
(cardiac)
Sardine gut fullness 3.0 (23)
(pyloric caeca)
Sardine $ gonad
3.7 (35)
stage
Sardine total length 196 (182225)
(mm)
Sardine total weight 62 (4594)
(g)
Sardine% positive
83.3
assays
Average anchovy
2.7 (025)
eggs per sh
Assayed sprat characteristics:
Sprat gut fullness
(cardiac)
Sprat gut fullness
(pyloric caeca)
Sprat total length
(average-range)
Sprat total weight
(average-range)
Sprat% positive
assays
Average anchovy
eggs per sh

1336.9

191 (167227) 188 (141208) 194 (176217) 193 (178217)

178 (138211) 179 (160213) 189 (164209)

187 (178198)

61 (3990)

57 (2083)

62 (4694)

62 (4581)

49 (2180)

46 (3276)

60 (3886)

55 (4671)

76.2

93.3

87.5

48.3

100

84.2

74.2

100

3.6 (025.7)

19.8 (080.6)

18.8 (0169)

2.5 (017.7)

41.8 (1151.7) 18 (061.8)

17.2 (0252.1)

2 (111.3)

2.2 (13)

2.1 (13)

3.0 (33)

3.0 (33)

1.9 (13)

2.1 (13)

3.0 (33)

3.0 (33)

3.0 (33)

99 (90105) 122 (112135)

88 (8098)

102 (93111)

106 (99117) 100 (93116) 121 (109133)

8 (611)

15 (1218)

6 (48)

9 (713)

10 (713)

9 (613)

17 (1322)

25

61.3

10

60

63.3

50

26.7

1.9 (028.5)

12.4 (070.9)

0.8 (04.4)

3.9 (049.2)

1.1 (027.2)

8.3 (051.2) 7.2 (042)

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Haul

95

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

(a)

100
90

% testing positive

80

y = 76.415e 4E -07x
R = 0.0437

70
60
50

y = 0.7098x 0.343
R = 0.3588

40
30
20
10
0
0

50000 100000 150000 200000 250000 300000 350000 400000 450000

Anchovy eggs (ind. 1000 m-2)

(b)

45

average predated eggs

(based in Ct)

40
35
30
25
20

y = 4.74e 3E -06x
R = 0.1429

15
10

y = 5.5833e -3E -06x

R = 0.2024

5
0
0

50000 100000 150000 200000 250000 300000 350000 400000 450000

Anchovy eggs (ind. 1000 m-2)

Fig. 7. Sardine and sprat predation on anchovy eggs. (a) Relationship between the occurrence of E. encrasicolus DNA remains in stomachs of S. pilchardus (black rhombus and
continuous line) and S. sprattus (white rhombus and broken line), the predation incidence, and the estimated abundance of E. encrasicolus eggs at the sh haul area (bottom
axis). Superimposed lines represent optimal regression model based on Akaike Information Criterion (AIC) tted to the data (respectively, P = 0.589 and P = 0.155 for sardine
and sprat curves). (b) Relationship between the average number of E. encrasicolus eggs predated by individual in each haul (minimum values based in assays Ct values; see
Section Mortality estimation due to sh predation) and the estimated abundance of E. encrasicolus eggs (bottom axis). Optimal regression model superimposed (respectively,
P = 0.316 and P = 0.311 for sardine and sprat).

DNA extraction and real-time PCR steps. In this sense, present ndings are not to be affected by this issue in a signicant way because
less than 1% of the controls (EBs and NTCs) presented a positive
signal pointing to pipetting malfunction. Beside this, PCR inhibitors
presence, potentially causing false negatives, has been reported
within DNA extracts from a wide range of sources including human

blood and soil (e.g. Rdstrm et al., 2004; Hedman and Rdstrm,
2013). Furthermore, PCR inhibition is a common issue when dealing with DNA extracts from stomach contents (e.g. Symondson,
2002; King et al., 2008) and has been reported when analyzing
stomach contents from a variety of crustaceans, including M. norvegica, and sh species (Albaina et al., 2010, 2012; Cleary et al.,

Table 5
Proportion of anchovy eggs mortality explained by sh predation (PC). Results are shown along with the data needed for computation. These includes data from the BIOMAN and
PELGAS 2010 surveys (Z and B respectively; ICES, 2010a,b) along with in situ produced data and others from available literature (see Section Mortality estimation due to sh
predation). Data are shown by geographical area for sardine but for the whole Bay of Biscay area for sprat due to the lack of regional acoustic biomass for the latter. For sardine,
acoustic biomasses from PELGAS 2010 campaigns south-offshore and Gironde areas were selected as they roughly corresponded to the two anchovy spawning centers in
2010.
Sardina pilchardus

Sprattus sprattus

Shelf-break

Gironde shelf

Bay of Biscay

t0 = duration of feeding (h)

g = gastric evacuation rate (h1)
EE = Average number of eggs per predator kg during time t0
C = Average anchovy eggs eaten daily per kg of predator
B = Biomass from acoustics (kg)
Ct = Total anchovy eggs eaten daily (C X B)
Gt = Total daily anchovy egg production
t = Anchovy egg development time (h)
Anchovy eggs daily instantaneous mortality rate
Z = Anchovy eggs hourly instantaneous mortality rate

24
0.2105
174.12
879.63
3.02E+07
2.66E+10
2.32E+12
72
0.34
0.0142

24
0.2105
567.31
2866.06
2.73E+07
7.83E+10
2.32E+12
72
0.34
0.0142

14
0.2105
538.47
1586.87
6.70E+07
1.06E+11
2.32E+12
72
0.34
0.0142

PC =

1.79E02

5.28E02

7.17E02

96

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

2012; Fox et al., 2012). Dilution of template DNA combined with

the addition of 1.25 lg/ll bovine serum albumin (BSA) to the
PCR reaction mix has been shown as effective in overcoming PCR
inhibition both in sh and marine invertebrates (Albaina et al.,
2010, 2012; Fox et al., 2012) and this was replicated here. Apart
from this, other biases associated with the assessment of predation, independently of the method applied, include regurgitation,
the risk of feeding within the sampling gear, scavenging and secondary predation. Regurgitation, a common issue in sh diet
assessment (e.g. Bowman, 1986), had a low if any incidence in
the present results as reected in the high stomach fullness
indexes reported for the assayed sh (Supp. Mat. Tables 3 and 4).
Moreover, the high sensibility and detectability of the real-time
PCR based method allows detecting feeding events even with
stomachs visually empty and thus reduces the potential affectation
by this issue (Albaina et al., 2010; Hunter et al., 2012). Feeding
within the sampling gear has been cited as a potential bias affecting zooplanktons predation rate estimations (Nicol, 1984); however, as expected due to the MIK net mesh size (1 mm ), no E.
encrasicolus egg (neither larva) was identied in the taxonomy
analysis thus avoiding this issue. Other factor to take into account
are alternative routes by which prey DNA might get into the guts of
predators which include secondary predation and scavenging (King
et al., 2008). While the secondary predation of anchovy eggs/larvae
by zooplankters would be highly unlikely and potentially undetectable by the assay, this probably does occur in sh predators
due to the combination of a higher predatory capacity and higher
detectability of DNA (e.g. Hunter et al., 2012). Apart from this,
scavenging would also imply an incorrect assignation of the egg/
larvae death cause but present method is not able to distinguish
this from predation and thus this issue cannot be dismissed
entirely.
Finally, for mortality estimation due to sh predation, we use
the Ct values of the real-time PCR assay as a proxy of the amounts
of eggs consumed per predator. As stated in Section Mortality estimation due to sh predation, to avoid the bias of the unknown
digestion time, we followed the most conservative approach calculating the minimum number of predated eggs that would yield the
observed Ct signal. This allowed us to compute the minimum percentage of anchovy eggs mortality that could be assigned to predation by sardines and sprats. The validity of our approach is
supported by the fact that the average number of anchovy eggs
estimated for sardine and sprat stomach contents (respectively,
15 and 5) fell within the range obtained by Bachiller (2013). It
has to be noted that traditional visual identication of egg remains
in the stomach contents is subjected to the same bias as the effect
of digestion time in the capacity to identify prey remains is also
ignored and cannot be compensated. Further developments combining quantitative PCR with fragment-length analysis may have
the potential to overcome the latter limitation (Deagle et al.,
2006; Troedsson et al., 2009). Apart from this, the real-time PCR
approach allows to screen hundreds-thousands of sh in a comparable and cost effective way (<1$ sample) without requiring any
taxonomical expertise. Regarding small crustacean, comprising
the bulk of zooplankton, molecular approaches stand as the only
option when studying predation on sh eggs where no hard
remains could resist the involved maceration.
Predation on anchovy ELS on May 2010
Predation incidence
A total of 32 taxa (out of 52 assayed) from the macrozooplankton community tested positive for the European anchovy
DNA assay indicating that the majority of the macrozooplankton
predator community was a potential predator on anchovy eggs at
the Gironde shelf spawning center. This is probably due to the

high concentration of eggs in the area and the resulting encounter rate. However, relatively high prey abundances were also
found in the shelf-break stations (Fig. 6) where predation was
low. In this sense, further studies should investigate the putative
role of alternative prey availability and/or species-specic feeding
behavior.
Regarding sh predators, although our survey covered a broad
range of prey abundances, high percentages of positive values for
anchovy DNA were found in every sardine haul with a minimum
predation incidence of 48% (haul 5) and the remaining 8 hauls well
over 70% incidence values (Table 4). Regarding sprat, although
positive signals were recorded in the seven hauls, predation incidence was lower than in sardine, ranging from 10% to 60% (Table 4).
Apart from this, both visual fullness and the molecular assay
results suggest that sprat have a diurnal feeding behavior while
sardines feed also during dark hours. This corresponds well with
the current knowledge on the physiology of the two species
(Garrido et al., 2007, 2008; Peck et al., 2012). This contrasting feeding behavior would increase the chances to eat anchovy eggs by
sardine as maximum anchovy spawning occurs at night (e.g.
Uriarte et al., 1996). The herein observed predation incidence
(82% for sardines and 42% for sprat) corresponds well with microscopic gut analysis where 55% and 30% of sardines and sprats presented anchovy eggs/larvae (Bachiller, 2013).

Estimations of egg mortality

In spatial terms our results suggest a high impact of the predation by macrozooplankton taxa on the spawning center at Gironde
shelf waters while this mortality at the shelf-break was considerably lower. The results indicate that a large fraction of anchovy
eggs mortality at the Gironde shelf spawning center is due to predation by macrozooplankton. This observation raises two points:
(1) Egg mortality is probably higher in shelf waters than in the
shelf-break. This has implications on the study of the effects of
transport in recruitment that has been generally directed to larvae
starvation but not to egg mortality (Allain et al., 2007a). And: (2)
Mortality due to macrozooplankton predation can be of the same
magnitude than that due to planktivorous sh predation. This
has been previously ignored when discussing potential factors
affecting Bay of Biscay anchovy recruitment (Irigoien et al.,
2007). It has also to be considered that gelatinous plankton has
not been included in this analysis. However, gelatinous organisms
are important predators of sh eggs and, for example, average daily
mortalities of >20% of eggs and larvae daily production has been
reported for gelatinous predators feeding on Anchoa mitchilli in
Chesapeake Bay (Purcell, 1985; Purcell et al., 1994; Purcell and
Arai, 2001).
Comparing to sardine, sprat showed lower predatory capacity
both in terms of anchovy eggs consumed per individual and of frequency of sprat individuals testing positive for anchovy DNA. This
could be counterbalance somehow by this species higher spatial
overlap with anchovy eggs at the Gironde shelf spawning center.
However, even considering the whole Bay of Biscay sprat biomass,
including the larger part of the stock located north of the anchovy
spawning centers (ICES, 2010b), the mortality does not exceed that
of sardine (7% for both species) that was computed with a more
precise predator biomass data (see Section Mortality estimation
due to sh predation). Regarding sardine, the herein reported PC
value (7%) is lower than the ones obtained by visual identication
of contents (1448%; Bachiller, 2013). However, this is probably
due to interannual differences in sardine biomass and overlap with
anchovy distribution. Finally, an important impact of cannibalism
has been reported for anchovy species (Hunter and Kimbrell,
1980; Alheit, 1987; Szeinfeld, 1991). Unfortunately, cannibalism
cannot be addressed with the herein developed method.

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

97

Concluding remarks

Appendix A. Supplementary material

The large range of macrozooplankton taxa assayed (52 taxa)

and the fact that, apart from anchovy, sardine and sprat comprised the bulk of planktivorous sh in the area (78%, ICES,
2010b), allows us to consider our results as a holistic view of
anchovy eggs predation mortality. It is noticeable that, at least
for the Gironde shelf spawning center, the observed mortality
due to macrozooplankton predation is in the same order of magnitude than the one computed for sh. Considered together, macrozooplankton and sh predation accounted for a large
percentage of the anchovy egg mortality. Apart from the previously neglected impact of macrozooplankton, the main nding
is the contrasting predation pressure that corresponded to the
two main spawning centers of Bay of Biscay anchovy. While relatively low eggs mortality rates, mainly due to sardine, were
recorded at the shelf-break spawning center, a higher predation
pressure was exerted by sardine, sprat and macrozooplankton
community at the Gironde shelf one.
Interestingly, reduced predation mortality at offshore environment has been proposed for anchovy juveniles. Irigoien et al.
(2007) suggested that anchovy juveniles could be recruiting
through a spatial loophole (sensu Bakun and Broad, 2003) for predation in these waters. Furthermore, a higher mortality on the Gironde shelf, compared with the offshore domain, was reported by
Cotano et al. (2008) when studying anchovy larval growth. Beside
this, present results suggest that the shelf-break spawning center
could be more favourable for anchovy eggs survival due to a lower
predation pressure. However, other factors such as differential
spawning intensity or advection patterns could be playing a role
(e.g. Allain et al., 2007a). In this sense, future studies could give
insights on the role of predation in the higher survival rate
reported for anchovy eggs produced after the peak spawning
moment (Allain et al., 2007b; Aldanondo et al., 2010). Related to
this, it has to be noted that the bulk of the taxa consuming anchovy
eggs/larvae in the Gironde shelf area belongs to the meroplankton
(Table 3; 67% of the positive signals). Because of this, the impact
exerted by this fraction of the macrozooplankton community, is
limited by the duration of both the predator planktonic stage and
the species-specic reproductive season. Wider temporal sampling
could assess the impact of macrozooplankton phenology on the
Bay of Biscay anchovy eggs/larvae survival.

Supplementary data associated with this article can be found, in

the online version, at http://dx.doi.org/10.1016/j.pocean.201
4.12.002.

Acknowledgements
We are grateful to the crews of the RV Investigador and RV
Emma Bardn and the BIOMAN 2010 campaign onboard scientists and analysts for their support during sampling. SGIker technical and human support (UPV/EHU) is gratefully acknowledged.
The authors are indebted to J. Illera and D. Prieto (UPV/EHU)
for their technical assistance and to J. Langa (UPV/EHU) for his
help on the statistical analyses of sh results. Thanks are due
to U. Laconcha (AZTI Tecnalia and UPV/EHU) for collecting and
preserving sh specimens for predation assessment, I. Zarraonaindia (UPV/EHU), M.A. Pardo and E. Jimnez (AZTI Tecnalia) and
P. Borsa (IRD, France) for providing European anchovy samples
for assay design and validation and, S. Milligan (CEFAS, UK) and
C. Fox (SAMS, UK) for providing clupeid samples for assay testing
purposes. This research was nancially supported by the projects
ECOGENBAY (MICINN CTM2009-13570-C02-02), funded by the
Ministry of Science and Research of the Government of Spain,
and BIOMAN, funded by the Department of Economic Development and Competitiveness of the Basque Government and by
the European Commission.

References
Albaina, A., Fox, C.J., Taylor, N., Hunter, E., Maillard, M., Taylor, M.I., 2010. A TaqMan
real-time PCR based assay targeting plaice (Pleuronectes platessa L.) DNA to
detect predation by the brown shrimp (Crangon crangon L.) and the shore crab
(Carcinus maenas L.)assay development and validation. Journal of
Experimental Marine Biology and Ecology 391, 178189.
Albaina, A., Taylor, M.I., Fox, C.J., 2012. Molecular detection of plaice remains in the
stomachs of potential predators on a atsh nursery ground. Marine Ecology
Progress Series 444, 223238.
Aldanondo, N., Cotano, U., Tiepolo, M., Boyra, G., Irigoien, X., 2010. Growth and
movement patterns of early juvenile European anchovy (Engraulis encrasicolus)
in the Bay of Biscay based on otolith microstructure and chemistry. Fisheries
Oceanography 19 (3), 196208.
Alheit, J., 1987. Egg cannibalism versus egg predation: their signicance in
anchovies. South African Journal of Marine Science 5, 467470.
Aljanabi, S.M., Martinez, I., 1997. Universal and rapid salt-extraction of high quality
genomic DNA for PCR based techniques. Nucleic Acids Research 25, 46924693.
Allain, G., Petitgas, P., Lazure, P., 2007a. The inuence of environment and spawning
distribution on the survival of anchovy (Engraulis encrasicolus) larvae in the Bay
of Biscay (NE Atlantic) investigated by biophysical simulations. Fisheries
Oceanography 16 (6), 506514.
Allain, G., Petitgas, P., Lazure, P., Grellier, P., 2007b. Biophysical modeling of larval
drift, growth and survival for the prediction of anchovy (Engraulis encrasicolus)
recruitment in the Bay of Biscay (NE Atlantic). Fisheries Oceanography 16 (6),
489505.
Andrs, M., Prellezo, R., 2012. Measuring the adaptability of eet segments to a
shing ban: the case of the Bay of Biscay anchovy shery. Aquatic Living
Resources 25, 205214.
Bachiller, E., 2012. Trophic studies of small pelagic sh in the Bay of Biscay:
methodological aspects. Revista de Investigacin Marina, AZTI-Tecnalia 19 (1),
120, <http://issuu.com/aztitecnalia/docs/rim_19_1/3?e=1899668/2473498>.
Bachiller, E., 2013. Trophic ecology of small pelagic sh in the Bay of Biscay:
ecological effects of trophic interactions. Ph.D. Thesis, University of the Basque
Country, Leioa, Spain, unpublished.
Baerwald, M.R., Schreier, B.M., Schumer, G., May, B., 2012. Detection of threatened
delta smelt in the Gut Contents of the invasive Mississippi Silverside in the San
Francisco Estuary Using TaqMan assays. Transactions of the American Fisheries
Society 141 (6), 16001607.
Bailey, K.M., Houde, E.D., 1989. Predation on eggs and larvae of marine shes and
the recruitment problem. Advances in Marine Biology 25, l83.
Bailey, K.M., 1994. Predation on juvenile atsh and recruitment variability.
Netherlands Journal of Sea Research 32, 175189.
Baker, R., Buckland, A., Sheaves, M., 2014. Fish gut content analysis: robust
measures of diet composition. Fish and Fisheries 15 (1), 170177.
Bakun, A., Broad, K., 2003. Environmental loopholes and sh population dynamics:
comparative pattern recognition with focus on El Nio effects in the Pacic.
Fisheries Oceanography 12, 458473.
Bax, N.J., 1998. The signicance and prediction of predation in marine sheries. ICES
Journal of Marine Science 55, 9971030.
Bellier, E., Planque, B., Petitgas, P., 2007. Historical uctuations in spawning location
of anchovy (Engraulis encrasicolus) and sardine (Sardina pilchardus) in the Bay of
Biscay during 196773 and 20002004. Fisheries Oceanography 16 (1), 115.
Bowles, E., Schulte, P.M., Tollit, D.J., Deagle, B.E., Trites, A.W., 2011. Proportion of
prey consumed can be determined from faecal DNA using real-time PCR.
Molecular Ecology Resources 11, 530540.
Bowman, R., 1986. Effect of regurgitation on stomach content data of marine shes.
Environmental Biology of Fishes 16, 171181.
Boyra, G., Rueda, L., Coombs, S.H., Sundby, S., dlandsvik, B., Santos, M., Uriarte, A.,
2003. Modelling the vertical distribution of eggs of anchovy (Engraulis
encrasicolus) and sardine (Sardina pilchardus). Fisheries Oceanography 12,
381395.
Bunn, N.A., Fox, C.J., Webb, T., 2000. A literature review of studies on sh egg
mortality: implications for the estimation of spawning stock biomass by the
annual egg production method. The Centre for Environment, Fisheries and
Aquaculture Science: Science Series Technical Report, 111. 37 pp.
Checkley, D.M., Ortner, P.B., Settle, L.R., Cummings, S.R., 1997. A continuous,
underway sh egg sampler. Fisheries Oceanography 6, 5873.
Cleary, A.C., Durbin, E.G., Rynearson, T.A., 2012. Krill feeding on sediment in the Gulf
of Maine (North Atlantic). Marine Ecology Progress Series 455, 157172.
Coombs, S.H., Boyra, G., Rueda, L.D., Uriarte, A., Santos, M., Conway, D.V.P., Halliday,
N.C., 2004. Buoyancy measurements and vertical distribution of eggs of sardine
(Sardina pilchardus) and anchovy (Engraulis encrasicolus). Marine Biology 145,
959970.
Coston-Clements, L., Wagget, R.J., Tester, P.A., 2009. Chaetognaths of the United
States South Atlantic Bight: distribution, abundance and potential interactions

98

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

with newly spawned larval sh. Journal of Experimental Marine Biology and
Ecology 373 (2), 111123.
Cotano, U., Irigoien, X., Etxebeste, E., Alvarez, P., Zarauz, L., Mader, J., Ferrer, L., 2008.
Distribution, growth and survival of anchovy larvae (Engraulis encrasicolus L.) in
relation to hydrodynamic and trophic environment in the Bay of Biscay. Journal
of Plankton Research 30 (4), 467481.
Deagle, B.E., Eveson, J.P., Jarman, S.N., 2006. Quantication of damage in DNA
recovered from highly degraded samplesa case study on DNA in faeces.
Frontiers in Zoology 3, 11.
Durbin, E.G., Casas, M.C., Rynearson, T.A., Smith, D.C., 2008. Measurement of
copepod predation on nauplii using qPCR of the cytochrome oxidase I gene.
Marine Biology 153, 699707.
Durbin, E.G., Casas, M.C., Rynearson, T.A., 2011. Copepod feeding and digestion rates
using prey DNA and qPCR. Journal of Plankton Research 34, 7282.
Folkvord, A., 1993. Prey recognition in stomachs of cannibalistic juvenile cod (Gadus
morhua L.). Sarsia 78, 97100.
Fox, C.J., Taylor, M.I., Pereyra, R., Villasana-Ortiz, M.I., Rico, C., 2005. TaqMan DNA
technology conrms likely over-estimation of cod (Gadus morhua L.) egg
abundance in the Irish Sea: implications for the assessment of the cod stock and
mapping of spawning areas using egg based methods. Molecular Ecology 14,
879884.
Fox, C.J., Taylor, M.I., Dickey-Collas, M., Fossum, P., Kraus, G., Rohlf, N., Munk, P., van
Damme, C.J.G., Bolle, L.J., Maxwell, D.L., Wright, P.J., 2008. Mapping the
spawning grounds of North Sea cod (Gadus morhua) by direct and indirect
means. Proceedings of the Royal Society of London, Series B: Biological Sciences
275, 15431548.
Fox, C.J., Taylor, M.I., van der Kooij, J., Taylor, N., Milligan, S.P., Albaina, A., Pascoal, S.,
Lallias, D., Maillard, M., Hunter, E., 2012. Identication of marine sh egg
predators using molecular probes. Marine Ecology Progress Series 462, 205
218.
Galluzzi, L., Penna, A., Bertozzini, E., Vila, M., Garces, E., Magnani, M., 2004.
Development of a real-time PCR assay for rapid detection and quantication of
Alexandrium minutum (a dinoagellate). Applied and Environmental
Microbiology 70, 11991206.
Garrido, S., Maralo, A., Zwolinski, J., Van der Lingen, C.D., 2007. Laboratory
investigations on the effect of prey size and concentration on the feeding
behaviour of Sardina pilchardus. Marine Ecology Progress Series 330, 189199.
Garrido, S., Ben-Hamadou, R., Oliveira, P.B., Cunha, M.E., Chcharo, M.A., Van der
Lingen, C.D., 2008. Diet and feeding intensity of sardine Sardina pilchardus:
correlation with satellite-derived chlorophyll data. Marine Ecology Progress
Series 354, 245256.
Goi, N., Logan, J., Arrizabalaga, H., Jarry, M., Lutcavage, M., 2011. Variability of
albacore (Thunnus alalunga) diet in the Northeast Atlantic and Mediterranean
Sea. Marine Biology 158 (5), 10571073.
Greenstone, M.H., Payton, M.E., Weber, D.C., Simmons, A.M., 2013. The detectability
half-life in arthropod predator-prey research: What it is, Why we need it, How
to measure it, and How to use it. Molecular Ecology, article rst published
online: 8 November 2013, doi: http://dx.doi.org/10.1111/mec.12552.
Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and
analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41,
9598.
Heath, M.R., 1992. Field investigations of the early life stages of marine sh.
Advances in Marine Biology 28, 1174.
Hedman, J., Rdstrm, P., 2013. Overcoming inhibition in real-time diagnostic PCR.
Methods in Molecular Biology 943, 1748.
Houde, E.D., 1987. Fish early life dynamics and recruitment variability. American
Fisheries Society Symposium 2, 1729.
Hunter, J.R., Kimbrell, C.A., 1980. Egg cannibalism in the northern anchovy, Engraulis
mordax. Fishery Bulletin 78, 811816.
Hunter, E., Taylor, N., Fox, C.J., Maillard, M., Taylor, M.I., 2012. Effectiveness of
TaqMan probes for detection of sh eggs and larvae in the stomach contents of a
teleost predator. Journal of Fish Biology 81 (1), 320328.
ICES, 2008. Report of the Workshop on Small Pelagics (Sardina pilchardus, Engraulis
encrasicolus) maturity stages (WKSPMAT), 1014 November 2008, Mazara del
Vallo, Italy. ICES CM 2008/ACOM:40. 82 pp.
ICES, 2010a. Report of the Working Group on Anchovy and Sardine (WGANSA), 24
28 June 2010, Vigo, Spain. ICES CM 2010/ACOM:16. 290 pp.
ICES, 2010b. Report of the Working Group on Acoustic and Egg Surveys for Sardine
and Anchovy in ICES Areas VIII and IX (WGACEGG), 2226 November 2010. ICES
CM 2010/SSGESST:24. 210 pp.
ICES, 2013. Report of the Working Group on Southern Horse Mackerel, Anchovy and
Sardine (WGHANSA), 2126 June 2013, Bilbao, Spain. ICES CM 2013/ACOM:16.
685 pp.
Ingle, R.W., 1991. Larval Stages of Northeastern Atlantic Crabs. An Illustrated Key.
Natural History Museum Publications, Chapman and Hall, London, 382 pp..
Irigoien, X., Fiksen, O., Cotano, U., Uriarte, A., Alvarez, P., Arrizabalaga, H., Boyra, G.,
Santos, M., Sagarminaga, Y., Otheguy, P., Etxebeste, E., Zarauz, L., Artetxe, I.,
Motos, L., 2007. Could Biscay Bay Anchovy recruit through a spatial loophole?
Progress in Oceanography 74, 132148.
Irigoien, X., De Roos, A., 2011. The role of intraguild predation in the population
dynamics of small pelagic sh. Marine Biology 158 (8), 16831690.
Jrme, M., Martinsohn, J.T., Ortega, D., Carreau, P., Verrez-Bagnis, V., Mouchel, O.,
2008. Toward sh and seafood traceability: anchovy species determination in
sh products by molecular markers and support through a public domain
database. Journal of Agricultural and Food Chemistry 56, 34603469.

Jumars, P.A., 2007. Habitat coupling by mid-latitude, subtidal, marine mysids:

import-subsidised omnivores. Oceanography and Marine Biology: An Annual
Review 45, 89138.
Jungbluth, M.J., Goetze, E., Lenz, P.H., 2013. Measuring copepod naupliar abundance
in a subtropical bay using quantitative PCR. Marine Biology 160, 31253141.
Kaartvedt, S., Larsen, T., Hjelmseth, K., Onsrud, M., 2002. Is the omnivorous krill
Meganyctiphanes norvegica primarily a selectively feeding carnivore? Marine
Ecology Progress Series 228, 193204.
King, R.A., Read, D.S., Traugott, M., Symondson, W.O.C., 2008. Molecular analysis of
predation: a review of best practice for DNA-based approaches. Molecular
Ecology 17, 947963.
Koster, F.W., Mollmann, C., 2000. Trophodynamic control by clupeid predators on
recruitment success in Baltic cod? ICES Journal of Marine Science 57, 310323.
Krautz, M.C., Gonzlez, M., Castro, L.R., 2003. Detection of anchoveta (Engraulis
ringens Jenyns 1842) eggs in euphausiid diets using immunoassays (ELISA).
Journal of Experimental Marine Biology and Ecology 294, 2739.
Krautz, M.C., Castro, L.R., Gonzlez, M., 2007. Interaction of two key pelagic species
in the Humboldt Current: euphausiid predation on anchoveta eggs estimated by
immunoassays. Marine Ecology Progress Series 335, 175185.
Lasker, R. (Ed.), 1985. An egg production method for estimating spawning biomass
of pelagic sh: application to the northern anchovy, Engraulis mordax. NOAA
Technical report NMFS 36, Dep Commer, Washington DC, US. 100 pp.
Lynam, C.P., Heath, M.R., Hay, S.J., Brierley, A.S., 2005. Evidence for impacts by
jellysh on North Sea herring recruitment. Marine Ecology Progress Series 298,
157167.
MacCall, A.D., 1981. The consequences of cannibalism in the stock-recruitment
relationship of planktivorous pelagic shes such as Engrauhs. In: Sharp, G.D.
(Ed.), Report and supporting documentation of the workshop on the effects of
environmental variation on the survival of larval pelagic shes, Lima,
1980.1.0.C. Workshop Rep. 28, Unesco, Paris, pp. 201220.
Mauchline, J., 1998. The biology of calanoid copepods. Advances in Marine Biology
33, 1710.
McBeath, A.J.A., Penston, M.J., Snow, M., Cook, P.F., Bricknell, I.R., Cunningham, C.O.,
2006. Development and application of real-time PCR for specic detection of
Lepeophtheirus salmonis and Caligus elongatus larvae in Scottish plankton
samples. Diseases of Aquatic Organisms 73, 141150.
Minegishi, Y., Yoshinaga, T., Aoyama, J., Tsukamoto, K., 2009. Species identication
of Anguilla japonica by real-time PCR based on a sequence detection system: a
practical application to eggs and larvae. ICES Journal of Marine Science 66,
19151918.
Molecular Ecology Resources Primer Development Consortium Abreu, A.G., Albaina,
A., Alpermann, T.J., Apkenas, V.E., et al., 2012. Permanent genetic resources
added to Molecular Ecology Resources Database 1 October 201130 November
2011. Molecular Ecology Resources 12, 374376.
Moser, H.G., Ahlstrom, E.H., 1985. Staging anchovy eggs. In: Lasker R. (Ed.), An egg
production method for estimating spawning biomass of pelagic sh:
Application to the northern anchovy, Engraulis mordax. NOAA Technical
Report NMFS 36, Dep Commer, Washington DC, US. pp. 3741.
Motoda, S., 1959. Devices of simple plankton apparatus. Memoirs of the Faculty of
Fisheries Hokkaido University 7, 7394.
Motos, L., 1994. Estimacin de la biomasa desovante de la poblacin de anchoa del
Golfo de Vizcaya, Engraulis encrasicolus, a partir de su produccin de huevos.
Bases metodolgicas y aplicacin. Ph.D. Thesis, University of the Basque
Country, Leioa, Spain, unpublished.
Motos, L., Uriarte, A., Valencia, V., 1996. The spawning environment of the Bay of
Biscay anchovy (Engraulis encrasicolus L.). Scientia Marina 60 (2), 117140.
Nejstgaard, J.C., Frischer, M.E., Simonelli, P., Troedsson, C., Brakel, M., Adiyaman, F.,
Sazhin, A.F., Artigas, L.F., 2008. Quantitative PCR to estimate copepod feeding.
Marine Biology 153, 565577.
Nicol, S., 1984. Cod end feeding by the euphausiid Meganyctiphanes norvegica.
Marine Biology 80, 2933.
Nikolioudakis, N., Palomera, I., Machias, A., Somarakis, S., 2011. Diel feeding
intensity and daily ration of the sardine Sardina pilchardus. Marine Ecology
Progress Series 437, 215228.
North, E.W., Gallego, A., Petitgas, P. (Eds.), 2009. Manual of recommended practices
for modelling physical biological interactions during sh early life. ICES
Cooperative Research Report 295. 111 pp.
Peck, M.A., Baumann, H., Bernreuther, M., Clemmesen, C., Herrmann, J.-P., Huwer, B.,
Kanstinger, P., Petereit, C., Temming, A., Voss, R., 2012. The ecophysiology of
Sprattus sprattus in the Baltic and North Seas. Progress in Oceanography 103,
4257.
Purcell, J.E., 1985. Predation on sh eggs and larvae by pe lagic cnidarians and
ctenophores. Bulletin of Marine Science 37, 739755.
Purcell, J.E., Nemazie, D.A., Dorsey, S.E., Houde, E.D., Gamble, J.C., 1994. Predation
mortality of bay anchovy (Anchoa mitchilli) eggs and larvae due to
scyphomedusae and ctenophores in Chesapeake Bay. Marine Ecology Progress
Series 114, 4758.
Purcell, J.E., Arai, M.N., 2001. Interactions of pelagic cnidarians and ctenophores
with sh: a review. Hydrobiologia 451, 2744.
Rdstrm, P., Knutsson, R., Wolffs, P., Lvenklev, M., Lfstrm, C., 2004. Pre-PCR
processing: strategies to generate PCR-compatible samples. Molecular
Biotechnology 26, 133146.
Redd, K.S., Ling, S.D., Frusher, S.D., Jarman, S., Johnson, C.R., 2014. Using molecular
prey detection to quantify rock lobster predation on barrens-forming sea
urchins. Molecular Ecology 23 (15), 38493869.

A. Albaina et al. / Progress in Oceanography 131 (2015) 8299

Richardson, D.E., Hare, J.A., Fogarty, M.J., Link, J.S., 2011. Role of egg predation by
haddock in the decline of an Atlantic herring population. Proceedings of the
National academy of Sciences of the United States of America 108, 13606
13611.
Santos, M., Uriarte, A., Ibaibarriaga, L., 2011. Spawning Stock Biomass estimates of
the Bay of Biscay anchovy (Engraulis encrasicolus L.) in 2010 applying the Daily
Egg Production Method. Revista de Investigacin Marina, AZTI-Tecnalia 18 (5),
7691.
Schooley, J.D., Karam, A.P., Kesner, B.R., Marsh, P.C., Pacey, C.A., Thornbrugh, D.J.,
2008. Detection of larval remains after consumption by shes. Transactions of
the American Fisheries Society 137, 10441049.
Somarakis, S., Palomera, I., Garca, A., Quintanilla, L., Koutsikopoulos, C., Uriarte, A.,
Motos, L., 2004. Daily egg production of anchovy in European waters. ICES
Journal of Marine Science 61, 944958.
Symondson, W.O.C., 2002. Molecular identication of prey in predator diets.
Molecular Ecology 11, 627641.
Szeinfeld, E., 1991. Cannibalism and intraguild predation in clupeoids. Marine
Ecology Progress Series 79, 1726.
Terazaki, M., 2005. Predation on anchovy larvae by a pelagic chaetognatha, Sagitta
nagae in the Sagami Bay, central Japan. Coastal Marine Science 29 (2), 162164.
ter Braak, C.J.F., Smilauer, P., 2002. CANOCO Reference Manual and CanoDraw for
Windows Users Guide: Software for Canonical Community Ordination (Version
4.5). Microcomputer Power.

99

Theilacker, G.H., Kimball, A.S., Trimmer, J.S., 1986. Use of an ELISPOT immunoassay
to detect euphausiid predation on larval anchovy. Marine Ecology Progress
Series 30, 127131.
Theilacker, G.H., Lo, N.C.H., Townsend, A.W., 1993. An immunochemical approach to
quantifying predation by euphausiids on the early stages of anchovy. Marine
Ecology Progress Series 92, 3550.
Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., 1997. The
CLUSTAL_X windows interface: exible strategies for multiple sequence
alignment aided by quality analysis tools. Nucleic Acids Research 25, 4876.
Tbe, K., Meyer, B., Fuentes, V., 2010. Detection of zooplankton items in the stomach
and gut content of larval krill, Euphausia superba, using a molecular approach.
Polar Biology 33, 407414.
Troedsson, C., Simonelli, P., Ngele, V., Nejstgaard, J.C., Frischer, M.E., 2009.
Quantication of copepod gut content by differential length amplication
quantitative PCR (dla-qPCR). Marine Biology 156, 253259.
Uriarte, A., Prouzet, P., Villamor, B., 1996. Bay of Biscay and Ibero Atlantic anchovy
populations and their sheries. Scientia Marina 60 (2), 237255.
Van Guelpen, L., Markle, D.F., Duggan, D.J., 1982. An evaluation of accuracy,
precision, and speed of several zooplankton sub-sampling techniques. Journal
du Conseil International pour lExplotation de la Mer 40, 226236.
Vestheim, H., Edvardsen, B., Kaartvedt, S., 2005. Assessing feeding of a carnivorous
copepod using species-specic PCR. Marine Biology 147, 381385.